WO2017112786A1 - Automatic suggestions for message exchange threads - Google Patents

Automatic suggestions for message exchange threads Download PDF

Info

Publication number
WO2017112786A1
WO2017112786A1 PCT/US2016/068059 US2016068059W WO2017112786A1 WO 2017112786 A1 WO2017112786 A1 WO 2017112786A1 US 2016068059 W US2016068059 W US 2016068059W WO 2017112786 A1 WO2017112786 A1 WO 2017112786A1
Authority
WO
WIPO (PCT)
Prior art keywords
user
conversation
suggestion
entity
message
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/US2016/068059
Other languages
English (en)
French (fr)
Inventor
Timothy Seeds Milligan
Hila Shemer
Dennis Kiilerich
Gang Ji
Ori Gershony
Sergey Nazarov
Pranav Khaitan
Sushant Prakash
Anton Volkov
Ricardo Escutia
David Citron
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Google LLC
Original Assignee
Google LLC
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Google LLC filed Critical Google LLC
Priority to EP16825663.4A priority Critical patent/EP3395018A1/en
Priority to JP2018551908A priority patent/JP6638087B2/ja
Priority to CN201680082643.2A priority patent/CN108781175B/zh
Priority to KR1020187019756A priority patent/KR20180093040A/ko
Publication of WO2017112786A1 publication Critical patent/WO2017112786A1/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
    • H04L51/21—Monitoring or handling of messages
    • H04L51/216—Handling conversation history, e.g. grouping of messages in sessions or threads
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
    • H04L51/02—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail using automatic reactions or user delegation, e.g. automatic replies or chatbot-generated messages
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/53—Network services using third party service providers
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
    • H04L51/04—Real-time or near real-time messaging, e.g. instant messaging [IM]
    • H04L51/046—Interoperability with other network applications or services
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L51/00—User-to-user messaging in packet-switching networks, transmitted according to store-and-forward or real-time protocols, e.g. e-mail
    • H04L51/06—Message adaptation to terminal or network requirements
    • H04L51/063—Content adaptation, e.g. replacement of unsuitable content

Definitions

  • the present disclosure relates to systems and methods for identifying an entity from a message exchange thread (also referred to herein as a "conversation"), and generating a suggestion for presentation to participant s) of the conversation via computing device(s) of the participant(s), where the suggestion is a suggestion for the participant(s) to take an action on the entity.
  • an entity is a thing or a concept that is singular, well- defined, and distinguishable.
  • an entity may be a person, place, item, idea, or any combination thereof.
  • the suggestion is directed toward the entity based on identifying the entity from one or more messages of the message exchange thread.
  • one or more contextual indicators of the entity are also determined, the contextual indicators are utilized to determine whether the entity is actionable, and the suggestion is provided for presentation to participant s) only when it is determined that the entity is actionable.
  • the entity may be determined to be actionable by determining whether the entity is, in a context of the message exchange thread, where a user is going to take an action on the entity.
  • Various contextual indicators may be utilized such as contextual indicators that are based on content external to the message exchange thread itself (e.g., prior electronic messages of participants in prior message exchange threads), contextual indicators that are based on natural language processing of message(s) (of the message exchange thread and/or prior message exchange thread(s)), contextual indicators that are based on
  • contextual indicators that are based on sentiment determined based on message(s) (of the message exchange thread and/or prior message exchange thread(s)), etc. Additional and alternative contextual indicators are described in more detail herein.
  • Providing suggestions only when it is determined (based on contextual indicators) that the entity is actionable may prevent participant(s) from being distracted by irrelevant suggestions and/or may conserve various computational resources by not transmitting and/or providing for presentation, irrelevant suggestions. For instance, suggestions may be provided only when they are determined to be contextually relevant, which, when compared to providing suggestions without regard to contextual relevance, may lead to a decrease in the providing of suggestions and a corresponding conservation of computational resources.
  • some implementations of providing suggestions according to techniques described herein may be of particular benefit to users that have low dexterity, or users who otherwise have difficulties making user inputs to a device.
  • the incorporation of the selectable suggestions into the message exchange thread may increase the reliability and speed of use of the device, e.g. by saving the user making errors due to low dexterity.
  • the suggestion is a selectable suggestion.
  • the suggestion in response to selection of the suggestion by a participant of the conversation, the suggestion (and/or related content) may be incorporated into a transcript of the conversation and/or a computer-based action directed toward the entity may be performed (e.g., within the message exchange thread and/or separate from the message exchange thread).
  • the suggestion may initially be provided as a non-static graphical element for presentation to one or more of the participants and, only upon selection of the suggestion by one or more of the participants, is the suggestion (and/or related content) statically incorporated into a transcript of the conversation.
  • the suggestion may form part of the transcript and continue to be viewed (e.g., by scrolling) in the conversation even after further messages have been submitted for inclusion in the transcript of the conversation.
  • the suggestion may be indexed in association with the conversation enabling participants to search for the conversation in the future based on the suggestion.
  • the suggestion may include an affirmative interface element and a negative interface element.
  • the suggestion may be incorporated in the transcript and in response to selection(s) of the affirmative interface element the suggestion may be discarded without static incorporation of the suggestion into the transcript.
  • providing contextually relevant suggestions during an ongoing message exchange thread may enable participants to add the suggestions (or related content) to the message exchange thread in a simple manner (e.g., via a single-tap and/or other single selection interface action).
  • contextually relevant suggestions may provide participant(s) with content that is directed to a contextually relevant entity, without requiring the participant(s) to switch from an application rendering the message exchange thread to another application in order to acquire such content.
  • providing a suggestion for presentation via a message exchange client and/or incorporating content based on the suggestion (e.g., in response to a user selection of the suggestion) into a transcript of a message exchange thread may obviate the need to switch from the message exchange client to acquire such content.
  • switching to another application may require usage of processor, memory, and/or battery resources via launching and/or surfacing of the application.
  • switching to another application to obtain content in lieu of obtaining the content in a message exchange thread may increase the amount of time a participant needs to spend obtaining the content - which may lead to a corresponding increase in consumption of computational resources in obtaining the content.
  • the participant(s) may be presented with contextually relevant suggestions, but still be allowed to control whether those suggestions form part of the conversation. This may mitigate clutter in user interfaces that render message exchange threads and/or may reduce usage of various computational resources (e.g., reduce storage space required to index or otherwise store a transcript of a message exchange thread since all suggestions are not automatically stored).
  • a method implemented by one or more processors includes receiving an electronic message submitted by a first user for display in a transcript of a message exchange thread.
  • the electronic message is submitted in response to user interface input provided by the first user via a client device of the first user.
  • the message exchange thread is between at least the first user and a second user via a first message exchange client of the first user and a second message exchange client of the second user.
  • the method further includes: identifying an entity from the electronic message, determining contextual indicators of the entity, and determining whether the entity is actionable based on the contextual indicators.
  • the method further includes, responsive to determining that the entity is actionable: providing, for presentation via at least one of the first message exchange client and the second message exchange client, a suggestion that is directed to the entity.
  • the suggestion is provided for presentation to the second user via the second message exchange client, but is not provided to the first user.
  • the method further includes: receiving an affirmative selection of the suggestion by the second user via the second message exchange client; and responsive to the affirmative selection, providing content based on the suggestion for presentation to the first user via the first message exchange client.
  • the content based on the suggestion is the suggestion itself.
  • the method further includes: receiving an affirmative selection of the suggestion; and responsive to the affirmative selection, incorporating content that is based on the suggestion in the transcript of the message exchange thread.
  • the method further includes: monitoring for selection of either the affirmative interface element or the negative interface element; and incorporating content that is based on the suggestion in the transcript of the message exchange thread only in response to selection of the affirmative interface element.
  • the method further includes: responsive to determining that the entity is actionable: transmitting the entity to a third-party server; and receiving the suggestion from the third-party server in response to transmitting the entity.
  • determining the contextual indicators includes determining at least one contextual indicator based on a separate prior message exchange thread that involves at least one of the first user and the second user.
  • the at least one contextual indicator is based on sentiment expressed by at least one of the first user and the second user in one or more prior messages of the prior message exchange thread.
  • the sentiment is directed to the entity and determining the at least one contextual indicator includes determining the at least one contextual indicator based on the sentiment being directed to the entity.
  • determining the contextual indicators includes determining at least one contextual indicator based on performing a sentiment analysis of a prior message of the first user or the second user.
  • the prior message is in the message exchange thread or in a prior message exchange thread.
  • performing the sentiment analysis is based at least in part on a non-textual graphical element included in the prior message.
  • the prior message is in the prior message exchange thread and the at least one contextual indicator is selected based on the sentiment being directed toward the entity in the prior message.
  • the suggestion is provided for presentation to the second user and the contextual indicators are based on a prior message of the first user.
  • the prior message is submitted by the first user in a prior message exchange thread that does not involve the second user.
  • the method further includes performing natural language analysis on the at least one electronic message based on a machine learning model, and identifying the entity that is actionable from the electronic message and determining the contextual indicators of the entity are based on the natural language analysis.
  • the method further includes tracking conversation flows from each participant of the message exchange thread to create tracking information, and generating the contextual indicators based on the tracking information.
  • the method further includes determining the suggestion using a trained machine learning model and the contextual indicators. In some of those implementations, the method further includes: receiving a user reaction to the suggestion provided for the entity; generating a training example with training example input based on the contextual indicators and with training example output based on the user reaction; and revising the trained machine learning model based on further training the trained machine learning model based on the training example.
  • the method further includes indexing, in a computer readable medium, the message exchange thread based on the contextual indicators.
  • implementations of the subject matter described in this disclosure may be embodied in methods that include receiving, using one or more processors, at least one conversation message from a conversation, identifying, using one or more processors, an entity that can be actionable from a conversation message, determining, using one or more processors, contextual indicators of the entity, determining, using one or more processors, whether the entity is actionable based on the contextual indicators, and responsive to the entity being actionable, providing a suggestion to a user to take an action on the entity.
  • the features may include performing a natural language analysis on the at least one conversation message based on a machine learning model and wherein identifying the entity that can be actionable from the conversation message and determining the contextual indicators of the entity are based on the analysis.
  • the features may include tracking conversation flows from each participant of the conversation and refining the contextual indicators based on the tracking information.
  • the features may include determining the suggestion for the user to take an action on based on a machine learning model and the contextual indicators.
  • the features may include receiving, from the user, an indication to mark the entity as actionable, sending the user indication as a training signal to a machine learning model, searching for information relevant to the entity and providing the information to the user.
  • the features may include receiving data from a variety of sources including queries and document retrievals, extracting features from the data, generating a machine learning model based on the extracted features, receiving a user action, the user action including a reaction to the suggestion provided for the entity, and training the machine learning module based on the user action.
  • the features may include feeding entity information to an application, the entity information including participants of the
  • the features may include detecting a question for the user in the conversation, the question being related to the entity, determining a potential user reply to the question, and providing the potential user reply in a one-tap form to the user.
  • the features may include organizing the at least one conversation based on the contextual indicators and indexing the at least one conversation.
  • a suggestion may be generated based on a machine learning model that learns over time to understand the semantics of a conversation and derives meaning from the real time conversation, and is therefore efficient and accurate.
  • the suggestion may also generated based on contextual analysis including sentiment analysis and conversation flows.
  • An appropriate suggestion is generated under an appropriate contextual condition, and is therefore intelligent. For example, no celebration message will be sent to a sad user.
  • the suggestion may offer rich context and next step(s) that can be directly used for a user to complete an action.
  • important suggestions and updates may be provided by other services in one place for all participants of a conversation.
  • the organization and summarization mechanism may allow users to easily look back at a meaningful trail of information shared in conversations.
  • Figure 1 is a block diagram illustrating an example system for identifying an entity from a conversation and generating a suggestion for a user to take an action on the entity.
  • Figure 2 is a block diagram illustrating an example suggestion application.
  • Figure 3 is a flowchart of an example method for identifying an entity from a conversation and generating a suggestion for a user to take an action on the entity.
  • Figure 4 is a flowchart of a specific example method for identifying an entity from a conversation and generating a suggestion for a user to take an action on the entity.
  • Figure 5 is a flowchart of an example method for manually determining an actionable entity from a conversation.
  • Figure 6 is a flowchart of an example method for generating and updating a machine learning model.
  • Figure 7 is a flowchart of an example method for feeding entity information from a first application to a second application to generate a suggestion.
  • Figure 8 is a flowchart of an example method for predicting a potential user reply in a conversation.
  • Figure 9 is a flowchart of an example method for selectively retrieving conversations and/or conversation messages in response to search requests.
  • Figure 10 includes graphic representations of example conversation interfaces showing a suggestion.
  • Figure 11 A is a graphic representation of an example conversation interface showing a conversation generated by a first application.
  • Figure 1 IB is a graphic representation of an example conversation interface showing a suggestion generated by a second application.
  • Figure 12 is a graphic representation of an example conversation interface showing one-tap replies.
  • Figure 13 is a graphic representation of an example conversation interface showing a conversation report.
  • Figure 14 is a graphic representation of an example conversation interface showing a conversation summary.
  • Figure 15 includes graphic representations of example conversation interfaces showing serial presentation of suggestions.
  • Systems, methods and interfaces for automatic suggestion generation are disclosed.
  • the systems, methods and interfaces for automatic suggestion generation are based on conversation analysis.
  • the message is analyzed to determine one or more entities that a user can take an action on.
  • Other messages of the conversation or even other conversations involving the user and/or other participant(s) of the conversation are then analyzed to determine the contextual indicators.
  • a suggestion is generated based on the contextual indicators. For example, if two users want to meet at a store, a map and a calendar could be suggested to direct the users to the store and set up a meeting time. Based on the contextual indicators, the map may only indicate directions to a particular store close to both users, and the calendar may highlight the available time of both users. In some instances, the suggestion offers next steps of the action.
  • the suggestion generation process is automatic in many aspects. For example, implementations of the present disclosure determine whether to generate a suggestion, and: when and where in the conversation to insert the suggestion; and/or when to recommend the suggestion to one or more participants for potential insertion in the conversation. For instance, for a question "meet at coffee house?" from user A to user B, a map that includes the coffee house may be suggested to one or both of the two users if user B answers "great! to the question, while the map suggestion will not be provided to either of the two users if user B answers "I'd like to go, but. .
  • different versions of the suggested map may be provided to each of the users, such as a first version provided to the first user with directions to the coffee house from the first user's current location and a second version provided to the second user with directions to the coffee house from the second user' s current location. Implementations of the present disclosure also determine what suggestion is appropriate, and/or when it is appropriate to provide the suggestion, based on contextual indicators. For example, if a user just received bad news (e.g., as determined based on analysis of another conversation of the user), it may be determined that a celebration suggestion is inappropriate for providing to the user in a conversation.
  • a user may be determined that a suggestion of such thing is not appropriate for providing to the user in the conversation.
  • client devices may provide some or all of the functionality described herein and may employ a cluster of servers in order to provide such functionality.
  • client hardware may be a mobile phone or tablet device.
  • Figure 1 illustrates a block diagram of a system
  • the illustrated system 100 includes client devices 102a, 102n that are accessed by users 104a, 104n, a server 106, an email server 108, a third-party server 1 10 and a search server 1 12. In the illustrated example, these entities are communicatively coupled via a network 1 14.
  • a letter after a reference number for example, " 102a” is a reference to the element having that particular reference number.
  • a reference number in the text without a following letter, for example, " 102" is a general reference to different instances of the element bearing that reference number.
  • the network 114 can be a conventional type, wired or wireless, and may have numerous different configurations including a star configuration, token ring configuration or other configurations. Furthermore, the network 114 may include a local area network (LAN), a wide area network (WAN) (e.g., the Internet), and/or other interconnected data paths across which multiple devices may communicate. In some instances, the network 114 may be a peer-to-peer network. The network 114 may also be coupled to or includes portions of a telecommunications network for sending data in a variety of different communication protocols.
  • LAN local area network
  • WAN wide area network
  • the network 114 may also be coupled to or includes portions of a telecommunications network for sending data in a variety of different communication protocols.
  • the network 114 includes Bluetooth communication networks or a cellular communications network for sending and receiving data including via short messaging service (SMS), multimedia messaging service (MMS), hypertext transfer protocol (HTTP), direct data connection, Wireless Application Protocol (WAP), email, etc.
  • SMS short messaging service
  • MMS multimedia messaging service
  • HTTP hypertext transfer protocol
  • WAP Wireless Application Protocol
  • Figure 1 illustrates one network 114 coupled to the client devices 102a, 102n, the email server 108, the third-party server 110 and the search server 112, in practice one or more networks 114 can be connected to these entities.
  • the client devices 102a, 102n in Figure 1 are used by way of example.
  • the users 104a, 104n interact with the client device 102a, 102n via signal line 116a, 116n respectively.
  • the client devices 102a, 102n are communicatively coupled to the network 114 via signal line 120a, 120n respectively, and exchange information with the server 106, the email sever 108, the third-party server 110 and the search server 112.
  • the client device 102a sends a conversation message (also referred to herein as an "electronic message") to the server 106.
  • the server 106 processes the conversation message and generates a suggestion based on the conversation message.
  • the server 106 receives one or more conversation messages and generates the suggestion based on the one or more conversation messages. In other instances, the server 106 also uses other inputs (e.g., prior user activities, user profile information, etc.) to generate the suggestion. Authorization is obtained prior to the use of any profile information or personal information of the user. The client device 102a receives the suggestion without unintended delays from the server 106.
  • the client device 102 receives a suggestion including an arrival time and a status of a flight right after sending a message "has John's flight landed?"
  • a delay may be desirable, for example, when a suggestion best fits at a future time in the conversation or when the server 106 waits for more conversation messages for context and confirmation to send the suggestion.
  • the server 106 may wait for a message from the client device 102n (e.g., "I hate coffee") before generating and/or providing the suggestion.
  • the client device 102 can be any computing device that includes a memory and a processor.
  • the client device 102 can be a laptop computer, a desktop computer, a tablet computer, a mobile telephone, a personal digital assistant, a mobile email device, a portable game player, a portable music player, a television with one or more processors embedded therein or coupled thereto or any other electronic device capable of accessing the network 114, etc.
  • the server 106 can be a hardware server that includes a processor, a memory and network communication capabilities.
  • the server 106 is communicatively coupled to the network 114 via signal line 122.
  • the server 106 sends and receives data to and from one or more of the client devices 102, the email server 108, the third-party server 110 and the search server 112 via the network 114.
  • the server 106 optionally includes a social network application 124.
  • a social network can be a network where the users may be connected by common feature(s).
  • the common feature(s) may include relationships/connections, e.g., friendship, family, work, an interest, etc.
  • the common feature(s) may be provided by one or more social networking systems including explicitly defined relationships and relationships implied by social connections with other online users, where the relationships form a social graph.
  • the social graph can reflect a mapping of these users and how they can be related.
  • the social network application 124 in the server 106 manages the social network by handling registration of users, publication of content (e.g. posts, comments, photos, links, check-ins, etc.), hosting multi-user communication sessions, managing of groups, managing different sharing levels, updating the social graph, etc.
  • publication of content e.g. posts, comments, photos, links, check-ins, etc.
  • the social network application 124 registers a user by receiving information such as a username and password and generates a user profile that is associated with the user and stored as part of the social graph.
  • the user profile includes additional information about the user including interests (e.g. soccer, reading, food, subscriptions, etc.), activities (e.g. search history, indications of approval, posts, comments, multi-user communication sessions, etc.), demographics (e.g. age, ethnicity, location, etc.) and profile rating and reputation (e.g., intelligence rating, humor rating, etc.).
  • the system 100 may include multiple servers 106 that include traditional social network servers, email servers, micro-blog servers, blog servers, forum servers, message servers, etc.
  • the server 106 and the social network application 124 may be representative of one social network.
  • a first social network may be more directed to business networking
  • a second may be more directed to or centered on academics
  • a third may be more directed to local business.
  • the email server 108 can be a hardware server that includes a processor, a memory and network communication capabilities.
  • the email server 108 is communicatively coupled to the network 1 14 via signal line 126.
  • the email server 108 exchanges user data with the server 106 so that the server 106 can generate a suggestion for an entity in a conversation message based on the user data. For example, when a user talks about buying a toy in a conversation, the server 106 can communicate with the email server 108 to get the user's purchase history (e.g., from purchase receipts for toys in emails), and determines a suggestion based at least in part on the purchase history. Authorization is obtained prior to the use of any purchase history or other personal information of the user.
  • the third-party server 1 10 can be a computing device that includes a processor, a memory and network communication capabilities.
  • the third-party server 1 10 is coupled to the network 1 14 via signal line 128.
  • the third-party server 1 10 sends and receives data to and from the client device 102, the server 106 and the search server 1 12 of the system 100 via the network 1 14.
  • the third-party server 1 10 may store different applications.
  • the server 106 can receive conversation messages from a conversation application stored on the third-party server 1 10.
  • the server 106 feeds entity information to an application stored on the third-party server 1 10 (e.g., via an application programming interface (API) of the third-party server 1 10) so that the application can generate a suggestion based on the entity information.
  • An application programming interface (API) of the third-party server 1 10) e.g., via an application programming interface (API) of the third-party server 1 10.
  • Multiple third-party servers may be provided and the server 106 may select an appropriate third-party server (for providing entity information and receiving a responsive suggestion) based on various factors. For example, a semantic type of the entity and/or action(s) mentioned in association with the entity in a conversation may be utilized to select a given third-party server in lieu of other available third-party servers.
  • third-party server 1 10 may provide suggestions for restaurant reservations and may be selected based on an entity being of a "restaurant” semantic type and/or based on "reserve", "book”, and/or other "reservation” terms being mentioned in association with the entity.
  • contextual indicators described herein may be utilized to determine whether an entity is actionable, and the server 106 may optionally provide entity information to a third-party server only when it is determined the entity is actionable. This may reduce network traffic by preventing some transmissions of entity information for an entity to a third-party server when the entity is not deemed actionable.
  • the search server 112 can be a computing device that includes a processor, a memory and network communication capabilities.
  • the search server 112 is coupled to the network 114 via signal line 130.
  • the search server 112 includes a search engine that provides a search result to a user responsive to a search request from the user.
  • the server 106 communicates with the search server 112 to search for information related to an entity and generates a suggestion to take an action on the entity based on the information returned by the search server 112 in response to the searching.
  • the server 106 comprises a suggestion application 132b.
  • the suggestion application 132b may additionally or alternatively be stored on the client device 102.
  • the suggestion application 132b may be a thin-client application that includes part of the suggestion application 132b on the client device 102 and part of the suggestion application 132b on the server 106 to identify an entity from a conversation and generate a suggestion for a user to take an action on the entity.
  • the suggestion application 132b receives at least one conversation message from a conversation and identifies an entity from a conversation message.
  • An entity can be a portion of a conversation message, for example, a person or an object in the conversation message.
  • the suggestion application 132b parses a conversation message based on analyzing the grammatical constituents of the conversation message and identifies one or more entities from the parsed conversation message. For example, the suggestion application 132b identifies that the message "great weather! includes “great,” “weather” and ' ! based on parsing the message, and determines a first entity "great” and a second entity "weather” from the message.
  • the suggestion application 132b further extends the analysis to the entire conversation or even other conversations (e.g., previous conversations, conversations from a third-party application) to determine contextual indicators. For example, from the message "would you like to meet at ABC coffee store?" and user profile information, the suggestion application 132b determines that an entity "ABC coffee store" is near the user's office. Authorization is obtained prior to the use of any profile information or personal information of the user.
  • Other contextual indicators include sentiment indicators, conversation flows, tense of the message, recency of the message, the day and/or time at which the message was sent, the day and/or time associated with the entity, conversation metadata, etc.
  • the suggestion application 132b further determines whether an entity is actionable based on determining whether the entity is in a context of conversation where a user is going to take an action on the entity. For example, the suggestion application 132b determines the A entity in the message "I like A" is non- actionable, but determines that the A entity in the message "I am going to buy A” is actionable. The suggestion application 132b determines whether an entity is actionable based on contextual indicators including conversation flows, tense, sentiment indicators (e.g., based on an emotion symbol), verbs used in the message, whether a question is asked in the message, conversation metadata, etc. If user A asks user B "meet at XYZ?," the suggestion application 132b will only determine that the location "XYZ" is actionable after receiving user B's positive answer in the conversation flow.
  • the suggestion application 132b determines whether an entity is actionable based on determining whether the entity is in a context of conversation where a user is going to take an action on the entity.
  • the suggestion application 132b generates a suggestion for an actionable entity based on contextual indicators. For example, the suggestion application 132b uses the messages "I hate A restaurant,” “I like B restaurant,” "C restaurant is ok” from a first user as contextual indicators to determine a suggestion B when a second user invites the first user to go to dinner. For example, the suggestion may be directed to B restaurant and/or C restaurant, but not A restaurant, based on the corresponding sentiments expressed with respect to each of those entities in the messages. As described herein, the messages may be from the conversation in which the second user invites the first user to go to dinner and/or may be from other conversations, that may or may not include the second user.
  • the suggestion B determined based on the first user's messages may be provided for presentation to the first user and/or the second user (if appropriate authorization is obtained).
  • the suggestion B may be generated based on the first user's messages, and provided for presentation to the second user.
  • the suggestion based on the first user's messages may only be provided to the second user if it is determined the first user's messages are from conversation(s) that also included the second user.
  • the suggestion based on the first user' s messages may be provided to the second user even if one or more of the first user' s messages are from conversation(s) that do not include the second user.
  • the suggestion application 132b may feed entity information to a separate application (e.g., hosted by third-party server 1 10) and, in response, get a suggestion from the separate application.
  • Feeding entity information to a separate application may include transmitting the entity information to the separate application utilizing one or more networks, and optionally utilizing an API of the separate application.
  • Receiving a suggestion in response to feeding the entity information may include receiving the suggestion over one or more networks in response to transmitting the entity information, and may optionally utilize an API (e.g., the same API utilized in transmitting the entity information).
  • a weather service may send an hourly weather report to the users.
  • the hourly weather report may be incorporated into a transcript of the conversation for viewing by all of the users and/or may be provided to the users via separate applications than the message exchange clients / applications utilized to engage in the conversation.
  • the suggestion application 132b connects users with the separate application without the users manually installing the separate application on a client device and provides important suggestions in a place for presentation to all participants of a conversation (e.g., in a transcript of the conversation).
  • the suggestion application 132b additionally or alternatively predicts a possible reply for a user to answer a question in a conversation, and so saves the user time. In other instances, the suggestion application 132b additionally or alternatively organizes information of a conversation so that users can easily reference back to any information they want in the conversation. Various implementations of the suggestion application 132b are discussed in further detail herein, e.g. with reference to Figure 2.
  • FIG. 2 is a block diagram of a computing device 200 that includes a processor 202, memory 204, a communication unit 208, a storage 210 and a suggestion application 132b according to some implementations.
  • the computing device 200 is a server 106.
  • the computing device is a client device 102.
  • the computing device is a distributed computing device such as one that operates on both server 106 and client device 102.
  • the components of the computing device 200 are illustrated in Figure 2 as communicatively coupled by a bus 206.
  • the processor 202 includes some or all of an arithmetic logic unit, a microprocessor, a general purpose controller or some other processor array to perform computations and provide electronic display signals to a display device.
  • the processor 202 is coupled to the bus 206 for communication with the other components.
  • Processor 202 processes data signals and may include various computing architectures including a complex instruction set computer (CISC) architecture, a reduced instruction set computer (RISC) architecture, or an architecture implementing a combination of instruction sets.
  • CISC complex instruction set computer
  • RISC reduced instruction set computer
  • Figure 2 includes a single processor, multiple processors may be included. The processing capability may be limited to supporting the display of images and the capture and
  • the processing capability might be enough to perform more complex tasks, including various types of feature extraction and sampling. It will be obvious to one skilled in the art that other processors, operating systems, sensors, displays and physical configurations are possible.
  • the memory 204 stores instructions and/or data that may be executed by processor 202.
  • the memory 204 is coupled to the bus 206 for communication with the other components.
  • the instructions and/or data may include code for performing any and/or all of the techniques described herein.
  • the memory 204 may be a dynamic random access memory (DRAM) device, a static random access memory (SRAM) device, flash memory or some other memory device known in the art.
  • DRAM dynamic random access memory
  • SRAM static random access memory
  • flash memory or some other memory device known in the art.
  • the memory 204 also includes a non-volatile memory or similar permanent storage and media, for example, a hard disk drive, a floppy disk drive, a CD-ROM device, a DVD-ROM device, a DVD-RAM device, a DVD-RW device, a flash memory device, or some other mass storage known in the art for storing information on a more permanent basis.
  • a non-volatile memory or similar permanent storage and media for example, a hard disk drive, a floppy disk drive, a CD-ROM device, a DVD-ROM device, a DVD-RAM device, a DVD-RW device, a flash memory device, or some other mass storage known in the art for storing information on a more permanent basis.
  • the communication unit 208 transmits and receives data to and from at least one of: the client devices 102, the server 106, the email server 108, the third-party server 110, and the search server 112.
  • the communication unit 208 is coupled to the bus 206.
  • the communication unit 208 receives data including a conversation message from the client device 102 and transmits the data to the server 106.
  • the server 106 generates a suggestion based on the conversation message using a suggestion application 132b stored on the server 106 and transmits the suggestion to the client device 102.
  • the communication unit 208 includes a port for direct physical connection to the client device 206 or to another communication channel.
  • the communication unit 208 includes an RJ45 port or similar port for wired communication with the client device 206.
  • the communication unit 208 includes a wireless transceiver (not shown) for exchanging data with the client device 206 or any other communication channel using one or more wireless communication methods, such as IEEE 1502.11, IEEE 1502.16, Bluetooth® or another suitable wireless communication method.
  • the communication unit 208 includes a cellular communications transceiver for sending and receiving data over a cellular communications network such as via short messaging service (SMS), multimedia messaging service (MMS), hypertext transfer protocol (HTTP), direct data connection, wireless application protocol (WAP), e-mail or another suitable type of electronic communication.
  • SMS short messaging service
  • MMS multimedia messaging service
  • HTTP hypertext transfer protocol
  • WAP wireless application protocol
  • the communication unit 208 includes a wired port and a wireless transceiver.
  • the communication unit 208 also provides other conventional connections to the network 114 for distribution of files and/or media objects using standard network protocols such as TCP/IP, HTTP, HTTPS and SMTP as will be understood to those skilled in the art.
  • the storage 210 is a non-transitory memory that stores data for providing the functionality described herein.
  • the storage 210 is coupled to the bus 206.
  • the storage 210 may be a dynamic random access memory (DRAM) device, a static random access memory (SRAM) device, flash memory or some other memory devices.
  • DRAM dynamic random access memory
  • SRAM static random access memory
  • the storage 210 also includes a non-volatile memory or similar permanent storage and media including a hard disk drive, a floppy disk drive, a CD-ROM device, a DVD-ROM device, a DVD-RAM device, a DVD-RW device, a flash memory device, or some other mass storage for storing information on a more permanent basis.
  • the storage 210 stores social network profiles associated with users in a social network, conversation messages in a conversation, actionable entities identified from the conversation, contextual indicators determined from the conversation and/or other conversations, suggestions generated based on the actionable entities of the conversation, etc.
  • the suggestion application 132b comprises a controller 220, a conversation management module 222, a model builder 224, a conversation analyzer 226, a determining engine 228, a suggestion generator 230, and a user interface engine 236.
  • the controller 220 can be code and routines for receiving data and transmitting the data to an appropriate component.
  • the controller 220 can be a set of instructions executable by the processor 202 to provide the functionality described below for receiving and transmitting the data.
  • the controller 220 can be stored in the memory 204 of the computing device 200 and can be accessible and executable by the processor 202.
  • the controller 220 can be adapted for cooperation and communication with the processor 202 and other components of the computing device 200.
  • the controller 220 receives data via the communication unit
  • the controller 220 receives user information from the client device 102 via the communication unit 208 and transmits the user information to the model builder 224 to build and update a machine learning model.
  • the controller 220 receives contextual indicators of an entity in a conversation from the conversation analyzer 226 and a determination that the entity is actionable from the determining engine 228, and transmits the contextual indicators of the actionable entity to the suggestion generator 230 to generate a suggestion for a user to take an action on the entity.
  • the conversation management module 222 can be code and routines for generating and managing a conversation session.
  • the conversation management module 222 can be a set of instructions executable by the processor 202 to provide the functionality described below for generating and managing a conversation session.
  • the conversation management module 222 can be stored in the memory 204 of the computing device 200 and can be accessible and executable by the processor 202.
  • the conversation management module 222 can be adapted for cooperation and communication with the processor 202 and other components of the computing device 200.
  • the conversation management module 222 generates a conversation session and instructs the user interface engine 236 to generate graphical data for displaying the conversation session in a conversation interface.
  • a conversation session or a conversation is a multi-user communication session that comprises at least one of a video conference, an audio conference, and instant messaging.
  • a conversation includes at least one conversation message.
  • a conversation message may include a video message, an audio message, a text message, a photo, a hyperlink, etc.
  • the conversation management module 222 generates a conversation by interacting with audio and video capture devices of the client device 102 to obtain a real-time audio-video synchronous communication data stream of the user 104.
  • the conversation management module 222 interfaces with a software driver stored on the client device 102 that controls the functionality of a microphone and a video camera (e.g., a webcam or forward facing camera) included in the client device 102, and generates a conversation based on the audio-video data captured by the client device 102.
  • the audio-video data can be encoded using various audio and video codecs.
  • the conversation management module 222 generates a conversation that allows users participating in the conversation to share documents, a scratchpad, and/or computing device screens with one another.
  • the conversation management module 222 receives a suggestion automatically generated by the suggestion generator 230 based on a conversation message of a conversation, and communicates with the user interface engine 236 to generate graphical data for displaying the suggestion along with the conversation message in the conversation.
  • the conversation management module 222 provides one or more applications or services in a conversation and instructs the user interface engine 236 to generate one or more tools in a conversation interface for a user to interact with the one or more applications or services.
  • the conversation management module 222 instructs the user interface engine 236 to generate a search box in a conversation interface for a user to search conversations based on topics, time, etc.
  • the conversation management module 222 also
  • the conversation management module 222 forwards the conversation messages to other components (e.g., 224-236) of the suggestion application 132b to generate a suggestion. In some instances, the conversation management module 222 also stores the conversation messages in storage 210.
  • the model builder 224 can be code and routines for generating and/or updating a machine learning model.
  • the model builder 224 can be a set of instructions executable by the processor 202 to provide the functionality described below for generating and/or updating a machine learning model.
  • the model builder 224 can be stored in the memory 204 of the computing device 200 and can be accessible and executable by the processor 202.
  • the model builder 224 can be adapted for cooperation and communication with the processor 202 and other components of the computing device 200.
  • the model builder 224 receives information from the conversation generated and received by the conversation management module 222, extracts features from the information and generates a model based on the extracted features.
  • the model includes items relevant to users such as a location, a list of interests, etc.
  • An interest can be explicit from a user's indication of what he or she likes.
  • An interest can also be information inferred from a user's activities. For example, if a user talks about hiking equipment in several
  • the model builder 224 infers that the user is interested in outdoor activities.
  • the model builder 224 also receives user information from a variety of sources including, for example, queries, document retrieval, user interactions, emails, etc., and builds a model based on the information.
  • the model builder 224 includes a user's search history into the model.
  • the model builder 224 upon receiving a collection of documents returned for a query and ranks associated with each document, the model builder 224 builds the model to include the ranks to show the relevancy between a document and the query.
  • the model builder 224 generates the model to include user statistics and user behavior pattern data, for example, how often a user joins a conversation, with whom a user talks most in conversations, what time a user likes to have a conversation with others, whether a user likes to share photos in a conversation, what kind of photos the user likes to share, etc.
  • the model builder 224 generates the model to include a sequence of weights and multipliers that are used to make predictions about a user action.
  • the model builder 224 determines weights and multipliers based on social relationships between users. For example, the model builder 224 receives conversation messages from the conversation management module 222.
  • the conversation messages include a question raised in a conversation and replies of a first user and a second user to the question.
  • the model builder 224 communicates with the social network application 124 to determine that the first user is a close friend of a third user and the second user does not know the third user personally based on the social graph managed by the social network application 124.
  • the model builder 224 assigns a higher weight to the first user than to the second user. Later when the model is used to predict the user's behavior based on the actions of similar users, the reply of the first user will weigh more in predicting a potential answer of the third user to the question. In other instances, the model builder 224 determines weights and multipliers based on recency of conversation messages.
  • the model builder 224 determines a higher weight for game B than game A when predicting which game forum the user would subscribe to. In some other instances, the model builder 224 determines weights and multipliers based on other factors including how often a user joins a conversation, what time a user likes to have a conversation with others, etc.
  • the model builder 224 generates a model and maps a user to one or more categories based on the model.
  • the category can be an entity identified by the conversation analyzer 226 as described below. For example, the model builder 224 determines that a user likes playing basketball, and categories the user into a category "sportsman" and a category "basketball.” Or the model builder 224 categories a user that talks with other users every day into a category "talker.”
  • the model builder 224 generates a machine learning model.
  • the model builder 224 collects data from a variety of sources including at least one conversation and uses a machine learning algorithm to learn certain properties from the collected data.
  • the machine learning algorithms may include, but are not limited to, regression techniques, instance-based algorithms, decision trees, Bayesian algorithms, kernel methods, clustering methods, association rule learning, artificial neural networks, etc.
  • the model builder 224 trains the model over time to understand the semantics of a conversation and derives meaning from the real time conversation. As the model evolves, the efficiency and accuracy of using the model in suggestion making, behavior prediction, etc., are improved.
  • the model builder 224 uses user actions as training signals to train the model.
  • the model builder 224 tracks actions of a user in previous conversations and trains the model to learn from those user actions. For example, the model builder 224 learns that a user never buys anything that was offered in a conversation interface with a purchase tool or learns that two users always choose a particular coffee house as their meeting spot from the previous conversations between them. In other instances, the model builder 224 tracks the non-chat actions of a user and trains the model to learn from the non- chat actions. For example, the model builder 224 learns a user's favorite chocolate brand from the user's purchase history. Based on what was learned by the model, a suggestion can be made for a user.
  • a suggestion of flight itineraries can be made to the user since the model builder 224 learns that the user likes this airline and often takes the flights around that time from previous actions of users.
  • the model builder 224 receives a user reaction to a suggestion provided to the user in a conversation and in turn uses the reaction (and other data from the conversation) as a training signal to train the model (e.g., uses the reaction to generate a training example that is used to refine the model through further training).
  • the model builder 224 may take this negative reaction as a signal to train the model. As a result, a different airline may be suggested to the user in the future.
  • the model builder 224 updates the model when receiving a request.
  • the suggestion generator 230 requests the model builder 224 to update the model when generating a suggestion using the model.
  • the model builder 224 periodically (once an hour, once a day, once a week, etc.) and/or non-periodically receives data and/or non-periodic data from a variety of sources for updating the model and updates the model accordingly.
  • the model builder 224 updates the model once a user action in a conversation, a non-chat action, or a user reaction is received.
  • the conversation analyzer 226 can be code and routines for analyzing a conversation to identify an entity that a user may take an action on from a conversation message and determine contextual indicators of the entity in the conversation.
  • the conversation analyzer 226 can be a set of instructions executable by the processor 202 to provide the functionality described below for analyzing a conversation to identify an entity that a user may take an action on from a conversation message and determine contextual indicators of the entity in the conversation.
  • the conversation analyzer 226 can be stored in the memory 204 of the computing device 200 and can be accessible and executable by the processor 202.
  • the conversation analyzer 226 can be adapted for cooperation and communication with the processor 202 and other components of the computing device 200.
  • the conversation analyzer 226 processes one or more conversation messages of a conversation received from the conversation management module 222 based on a machine learning model generated by the model builder 224.
  • the conversation analyzer 226 analyzes each conversation message using natural language processing (NLP) techniques.
  • NLP natural language processing
  • the conversation analyzer 226 parses the conversation message and implements NLP entity extraction on the parsed conversation message to obtain an entity.
  • the entity can be a portion of a conversation message, for example, a person or an object in the conversation and the entity can be at various levels of granularity (e.g., a restaurant, an Italian restaurant, a particular Italian restaurant).
  • the NLP entity extraction is based on a semantic annotation for text (SAFT) library that works in the machine learning model.
  • SAFT semantic annotation for text
  • the model builder 224 receives a flight number, builds an annotator into the model to recognize the flight number in a conversation based on the general format of flight numbers, and trains the annotator.
  • the conversation analyzer 226 communicates with the model builder 224 to use the annotator to determine that one entity in the conversation message is the flight number.
  • the conversation analyzer 226 parses and extracts entities "beautiful,” "weather” and "Store A" from a conversation message.
  • the conversation analyzer 226 determines that "beautiful” is an adjective, "weather” is a noun, and "Store A” is a location and a noun based on the language semantic analysis using the SAFT library.
  • An entity identified by the conversation analyzer 226 can be actionable or non-actionable depending on whether a user can take an action on it. The determination whether an entity is actionable or non-actionable is described in more detail with reference to the determining engine 228.
  • the conversation analyzer 226 determines contextual indicators of the entity in the
  • the conversation analyzer 226 determines when the location was mentioned in the conversation, who mentioned the location in the conversation, which content of the conversation included the location, which kind of content followed the location in the conversation, and/or other contextual indicators. In some instances, the conversation analyzer 226 analyzes multiple conversations (e.g., previous conversations, conversations generated by other applications) to determine contextual indicators of an entity in a conversation.
  • conversations e.g., previous conversations, conversations generated by other applications
  • the conversation analyzer 226 performs NLP to determine the contextual indicators of an entity in a conversation.
  • the NLP is based on the machine learning model generated by the model builder 224 and may be more sophisticated than conventional NLP in many aspects.
  • the conversation analyzer 226 performs the NLP based on pattern matching in combination with user information of one or more users in a conversation. For example, the conversation analyzer 226 can detect the phrase "Basil Thai restaurant” from a conversation message based on pattern matching, and match that phrase to one or more particular restaurants named "Basil Thai" based on user information of one or more user(s) (e.g.
  • the conversation analyzer 226 can detect a particular "Coffee house” near user A from a conversation message based on pattern matching in combination with user information.
  • using user information to select one or more particular entities may lead to improved suggestions. For example, based on “Basil Thai restaurant” without user information, 3 restaurants named “Basil Thai” may be identified and, if a driving direction is suggested to participant(s) in a conversation, the suggestion may include three maps to all three "Basil Thai restaurants". The participant s) then have to choose which map is relevant and accurate.
  • only one relevant and accurate driving direction to a "Basil Thai restaurant” may be provided (e.g., the one that is most convenient to all participants of the conversation and/or the one most frequently visited by participant(s)).
  • the conversation analyzer 226 additionally or alternatively determines contextual indicators by performing sentiment analysis on a conversation message and associating the conversation message with a weight. For example, the conversation analyzer 226 receives messages "I hate coffee,” “tea is ok” and “coke is my favorite” from user A, and associates a most favorable to least favorable sentiment weight to the messages including coke, tea, and coffee, respectively.
  • the weights indicate user A' s beverage taste. In other words, the weights may each indicate a sentiment of user A with respect to each "beverage” entity corresponding to the entities of "coffee", "tea", and "coke”.
  • the messages may be from a current conversation in which a suggestion is to be provided and/or from a prior conversation that includes the user (with or without other participants of the current conversation).
  • the conversation analyzer 226 assigns a positive number to indicate a positive sentiment, a negative number to indicate a negative sentiment, and a zero to indicate indifference.
  • the conversation analyzer 226 may use the magnitude of the number as the weight or intensity of a sentiment and may assign the sentiment to a corresponding entity. For the above example, the conversation analyzer 226 weights a positive two, a positive one, and a negative two, to the messages including coke, tea and coffee, respectively.
  • the conversation analyzer 226 determines a sentiment from an emotion symbol and/or other non-textual element included in a conversation message and assigns a weight to the sentiment. For example, the conversation analyzer 226 may determine that a user is happy and excited from the message "mom is coming ⁇ ," (e.g., based on presence of the emotion symbol " ⁇ ” and assignment of that symbol to a happy sentiment) and may determine that the user is unhappy from the message "I wish mom is still here ⁇ " (e.g., based on presence of the emotion symbol " ⁇ " and assignment of that symbol to a sad sentiment) The conversation analyzer 226 may associate each message with a positive number (or other measure) or a negative number (or other measure) to indicate how happy the user is.
  • a positive number or other measure
  • a negative number or other measure
  • the conversation analyzer 226 may determine an overall sentiment based on the weights and the number of the conversation messages. Such information become contextual indicators and contribute to generating a suggestion. For example, if a user says that "I wish mom is still here ⁇ ,” “Amy is not here too” and "I should go see a movie,” the conversation analyzer 226 determines that overall the user is not happy. As a result, a comedy movie instead of a tragedy movie may be suggested to cheer the user up. The sentiment analysis on a conversation therefore makes a suggestion more intelligent.
  • the conversation analyzer 226 determines contextual indicators by taking more than one piece of conversation message for analysis.
  • Standard and conventional conversational semantics may be generated, for example, “Barack Obama” in response to a query of "Who is the president of the USA?" in a conversation and to generate “Michelle Obama” in response to a query of "Who is the wife of Barack Obama?" in the conversation. Later when "him” and "her” are mentioned in the conversation, the conversation analyzer 226 references him with “Barack Obama” and references her with “Michelle Obama” using coreference resolution.
  • the conversational LP may be similar to conventional conversational semantics in some implementations. Both conventional conversational semantics and the conversational NLP may be based on semantic analysis on individual messages.
  • the conversation analyzer 226 may further analyze contiguous and/or non-contiguous messages in a conversation rather than analyzing only a single message.
  • the conversation analyzer 226 analyzes multiple conversation messages to track conversation flows from each participant of a conversation.
  • the conversation analyzer 226 tracks users for possible participants of a conversation and determines the conversation flow, i.e., which participant sends which message in what order.
  • the conversation analyzer 226 refines the contextual indicators based on the tracking information. For example, the conversation analyzer 226 may analyze two contiguous messages "would you like to meet?" and "how about ABC?" of a conversation.
  • the conversation analyzer 226 tracks participants of the conversations, and determines that a first participant sent the first message and a second participant sent the second message.
  • the conversation analyzer 226 determines that the location entity "ABC" is proposed by the second participant to respond to a question from the first participant.
  • the conversation analyzer 226 refines the contextual indicators based on information tracked from multiple conversations.
  • the conversation analyzer 226 may refine the contextual indicators based on common terms that have occurred in multiple conversations. For example, the conversation analyzer 226 identifies an entity "hiking" from a message “do you want to go hiking?" sent from user A when talking with user B today.
  • the conversation analyzer 226 tracks conversation flows and determines that a message "I prefer fishing to hiking" from user B in a previous conversation should be used as contextual indicators for the entity "hiking.”
  • conversation analyzer 226 may refine the contextual indicators based on concept clustering of terms occurred in multiple conversations. For example, the conversation analyzer 226 determines messages including "mountain bike,” “backpack,” “camping” from multiple conversations can be used as contextual indicators for the message “do you want to go hiking” received in an ongoing conversation since the terms “mountain bike,” “backpack,” “camping” and “hiking” can be classified into a cluster "outdoor activity” based on concept clustering.
  • the conversation analyzer 226 also analyzes a conversation to determine the tense (e.g., whether a conversation message is in the past tense), the recency (e.g., whether a conversation message is new, how to weigh a new or old conversation message), etc.
  • the conversation analyzer 226 further collects data including user profile information (e.g., an age, a gender), conversation metadata (e.g., a start time and an end time of a conversation, participants of the conversation, timestamps associated with each conversation message), and uses the collected data as contextual indicators. Authorization is obtained prior to the use of any profile information or personal information of the user.
  • the determining engine 228 can be code and routines for determining whether an entity is actionable.
  • the determining engine 228 can be a set of instructions executable by the processor 202 to provide the functionality described herein for determining whether an entity is actionable.
  • the determining engine 228 can be stored in the memory 204 of the computing device 200 and can be accessible and executable by the processor 202.
  • the determining engine 228 can be adapted for cooperation and communication with the processor 202 and other components of the computing device 200.
  • the determining engine 228 determines whether the determined entity is actionable. An entity is actionable if a user is likely to take an action on it. The determining engine 228 determines an entity is an actionable entity when the entity is in a context of conversation where a user is likely to take an action on the entity. As described herein, whether a user is likely to take action on an entity is determined by the determining engine 228 utilizing one or more of various techniques and is based on one or more contextual parameters described herein. In some instances, the determining engine 228 detects one or more action verbs, the subject and/or the object of the one or more verbs and associated context, and determines an actionable entity based on the detected information.
  • An action verb is a verb that expresses physical or mental action such as walk, read, want, think, etc.
  • the conversation analyzer 226 identifies the location entity "supermarket C" from a message “meet at supermarket C” between user A and user B in the conversation
  • the determining engine 228 may detect an action verb "meet,” an object “supermarket C” of the verb "meet,” and determine that this location entity "supermarket C” is actionable because users A and B may meet at the place.
  • the conversation analyzer 226 identifies the location entity "supermarket C" because user A comments that "supermarket C is super good” in the conversation, the determining engine 228 fails to detect an action verb and associated object/subject.
  • the determining engine 228 may not determine that the location entity "supermarket C" is actionable because no action is detected.
  • the determining engine 228 classifies the entity as actionable or non-actionable.
  • the determining engine 228 highlights or otherwise annotates the actionable entity for use by one or more other components as described herein.
  • the conversation analyzer 226 analyzes all conversation messages in the conversation to determine contextual indicators of an entity identified from a conversation message of the conversation.
  • the determining engine 228 determines whether the entity is actionable based on the contextual indicators. In some instances, the determining engine 228 determines whether an entity is actionable based on a conversation flow. For example, user A enters a question "meet at ABC?" In certain examples, the determining engine 228 may not determine whether the location "ABC" is actionable until receiving user B's answer in the conversation flow. If user B says “Sure,” the determining engine 228 then determines that "ABC" is actionable. Otherwise, the determining engine may determine "ABC" as non-actionable.
  • the determining engine 228 may determine "DEF” rather than "ABC” as an actionable entity. In other instances, the determining engine 228 determines whether an entity is actionable based on the tense of the message. For example, the determining engine 228 may determine that the flight number 1234 in the message "I used to take 1234 to New York" is non-actionable since the message is in the past tense.
  • the determining engine 228 determines whether an entity is actionable based on sentiment indicators. For example, the determining engine 228 determines "pasta" is non- actionable based on the message "want pasta for dinner?" from user A and the message "I hate pasta” from user B. However, the determining engine 228 may still determine that "dinner” is actionable. A suggestion that gives more details or completes an action (e.g., a pizza restaurant) may still be provided to user A. The determining engine 228 also determines whether an entity is actionable based on other contextual indicators determined by the conversation analyzer 226.
  • the determining engine 228 determines whether an entity is actionable based on contextual indicators, which differs the present disclosure from some approaches and may therefore be advantageous.
  • the conventional approaches may determine that today is Alice's birthday based on a previous conversation between Alice and her friend, and automatically send her a "Happy Birthday" message and a fun video clip to celebrate her birthday. Implementations of the present disclosure sends such automatic celebration messages to Alice only if the contextual condition is determined to be appropriate. If Alice said that "my house was burn down" a few hours ago, the determining engine 228 decides not to send any celebration message to Alice since the message can only be an annoying spam for Alice.
  • a suggestion directed to "Coffee Shop A” may be provided to User B for incorporation in the current conversation.
  • the suggestion may be provided based on User A's previous expression of positive sentiment directed toward "Coffee Shop A” and may optionally be provided even if User B was not involved in the prior conversation.
  • the determining engine 228 in addition to automatically determining an actionable entity based on contextual indicators in a conversation, the determining engine 228 also allows a user to manually determine which entity is actionable. Even if the determining engine 228 did not detect that an entity is actionable, the determining engine 228 allows a user to make the entity actionable. As a result, any message in a conversation can be actionable. In some instances, the determining engine 228 receives an indication from a user to mark an entity in a conversation message as actionable and returns information relevant to the entity to the user. For example, a user mentions "panda" in a conversation message.
  • the user determines that this word is the most important entity in the message and indicates that this word is actionable (e.g., by labelling the word via user interface input such as a "tap on” or other selection of the word). Responsive to receiving this indication, the determining engine 228 communicates with other servers or applications (e.g., a search server 112 and/or third-party server 110) to retrieve information related to "panda" (e.g., a first search result from search server 112) to the user. In some instances, responsive to receiving an indication that a user wants to take an action on an entity, the determining engine 228 also sends the indication as a training signal to the model builder 224 to train the machine learning model based on the indication.
  • a search server 112 and/or third-party server 110 to retrieve information related to "panda" (e.g., a first search result from search server 112) to the user.
  • the determining engine 228 responsive to receiving an indication that a user wants to take an action on an entity
  • the model builder 224 may, based on the indication, generate a positive training example with training example output indicating an actionable entity and training example input that is based on the message in which the indication was made, the word(s) that were the focus of the indication, and/or contextual indicators associated with the conversation.
  • the model builder 224 may refine the machine learning model by further training the model based on the positive training example.
  • the suggestion generator 230 can be code and routines for generating a suggestion for a user to take an action on an entity.
  • the suggestion generator 230 can be a set of instructions executable by the processor 202 to provide the functionality described herein for generating a suggestion for a user to take an action on an entity.
  • the suggestion generator 230 can be stored in the memory 204 of the computing device 200 and can be accessible and executable by the processor 202.
  • the suggestion generator 230 can be adapted for cooperation and
  • the suggestion generator 230 Responsive to receiving an actionable entity from the determining engine 228, the suggestion generator 230 generates a suggestion for the actionable entity to offer richer context and next steps.
  • the suggestion is information that is useful to make an entity more actionable for a user or to complete an action for a user. For example, when users suggest meeting at a restaurant, the suggestion generator 230 can provide more details about the restaurant and offer to book a table. When users talk about watching a movie, the suggestion generator 230 can offer show times or a link to the movie stream. When users mention a video clip from a website, the suggestion generator 230 can bring it up for users to play in the conversation. In some instances, the suggestion generator 230 generates a real-time suggestion. For example, when users are talking about lunch at a restaurant, the suggestion generator 230 generates a map in real time to direct the users to the restaurant.
  • the suggestion generator 230 instructs the user interface engine 236 to incorporate at least one suggestion tool into a conversation interface.
  • the suggestion generator 230 uses the suggestion tool to display a suggestion and receive user interactions with the suggestion.
  • the suggestion generator 230 instructs the user interface engine 236 to generate at least one suggestion tool to display a map for a user viewing and selecting, to display a calendar for a user viewing and scheduling, to find and share contact information, to share a file, to set up a reminder, to display a video or initiate a video call, to share stickers, to display weather information, to display sports scores, to display product or service (e.g., clothes, movies, medical service) purchase information, to display flight information, to provide traffic data, to provide local event information, to provide local merchant information (e.g., restaurants, movie theaters), to share user status (e.g., driving, biking, running), to find and share a product (e.g., a mobile application, a game), to pay or receive a payment for a product or
  • a product
  • the suggestion provided by the suggestion generator is a suggestion provided by the suggestion generator
  • the suggestion generator in response to selection of the suggestion by a participant of the conversation, may incorporate the suggestion (and/or related content) into a transcript of the conversation.
  • the suggestion generator 230 in response to selection of the suggestion, may cause it (and/or related content) to be statically displayed in the transcript for all participants of the conversation.
  • the suggestion may initially be provided to the user interface engine 236 for presentation to one or more of the recipients in their corresponding conversation interfaces as a non-static graphical element (e.g., presented in the interface, but not fully incorporated in the transcript).
  • the suggestion in response to selection of the suggestion by one or more of the participants, the suggestion (and/or related content) may be statically incorporated into the transcript of the conversation.
  • the suggestion may include an affirmative interface element and a negative interface element.
  • the suggestion In response to selection(s) of the affirmative interface element, the suggestion may be incorporated in the transcript and in response to selection(s) of the affirmative interface element the suggestion may be discarded without static incorporation of the suggestion into the transcript.
  • the suggestion generator 230 generates a suggestion based on what the machine learning model from the model builder 224 learns from user actions.
  • the model builder 224 trains the machine learning model to learn from user' s non-chat actions and the suggestion generator 230 generates a suggestion based on the machine learning model. For example, if a user talks about taking a flight to Las Vegas next weekend, the determining engine 228 determines an actionable entity "flight" and the suggestion generator 230 offers a suggestion of a flight itinerary (e.g., time and airline) based on user's previous purchase history learned in the machine learning model.
  • a flight itinerary e.g., time and airline
  • the model builder 224 trains the machine learning model to learn from reactions of a user to previous suggestions and the suggestion generator 230 generates a suggestion based on the machine learning model.
  • the suggestion generator 230 may communicate with the user interface engine 236 to customize a new suggestion tool in a conversation interface based on user actions taken to previous suggestion tools offered in the conversation interface. For example, the model builder 224 trains the machine learning model to learn that a user never made any purchase using purchase tools previously offered in a conversation interface displayed to the user. As a result, the suggestion generator 230 determines no longer to incorporate a purchase tool in the conversation interface.
  • the suggestion generator 230 used to provide a traffic map tool and a public transport tool to a user.
  • the suggestion generator 230 learns from the machine learning model that the user always interacted with the public transport tool, and therefore determines to generate and provide only a public transport tool to the user in the future.
  • user A and user B always discuss or select Coffee house as a meetup spot.
  • the suggestion generator 230 generates a map suggestion and instructs the user interface engine 236 to include a map tool and highlight the Coffee house nearby in the map tool.
  • the suggestion generator 230 generates a suggestion based on the conversation analysis (e.g., LP) and contextual indicators determined from the conversation analyzer 226.
  • the suggestion generator 230 generates a suggestion based on the sentiment analysis.
  • the suggestion generator 230 may highlight the available time around noon on the calendar based on the sentiment analysis.
  • the suggestion generator 230 generates a suggestion based on user information conveyed in a conversation. For example, if user A says “let' s meet tomorrow" and user B says “great! I'll be in San Francisco,” the suggestion generator 230 may provide a map of San Francisco to users A and B even if neither user is at San Francisco when the conversation occurs.
  • the suggestion generator 230 generates a suggestion based on user profile information.
  • the suggestion generator 230 generates a suggestion by pulling data from a search engine and/or other applications into a conversation.
  • the suggestion generator 230 receives map data around a location from a map service in the conversation and generates a map suggestion for the actionable location entity.
  • An example conversation interface including a suggestion is depicted in Figure 10.
  • the suggestion generator 230 instead of itself generating a suggestion for a user to take an action on an entity, the suggestion generator 230 feeds the entity information into other applications and services.
  • the other applications and services may themselves optionally directly provide a suggestion for a user to take an action on the entity. For example, if a flight number is mentioned in a conversation, the suggestion generator 230 notifies a third-party server to provide flight feeds (e.g., delay information) to one or more participants in the conversation (e.g., by
  • the suggestion generator 230 communicates with an advertisement application to use this information to target the user with advertising. If the user mentions a sports game in the conversation, the suggestion generator 230 communicates with a third-party service to provide the sports scores of the sports game to the user.
  • An example conversation interface including a suggestion provided by a different application is depicted in Figures 1 1 A and 1 1B.
  • the entity information includes participants of the conversation from which the entity is identified. When the suggestion generator 230 feeds the entity information into other services, the other services may provide a suggestion to all (or a subset of) the participants of the conversation.
  • a weather service may send a hourly weather report to users A, B and C after receiving entity information of "weather" from the suggestion generator 230.
  • the suggestion generator 230 also communicates with the conversation management module 222 and the user interface engine 236 to initiate a live conversation in a conversation interface.
  • the other service provides a suggestion for the user to take an action on the entity in the live conversation.
  • the other services may provide the hourly weather report either in an interface provided by the other services or in the live conversation provided by the conversation management module 222 and the user interface engine 236.
  • the suggestion generator 230 can provide smarter, more relevant, more searchable suggestions to a user, and therefore is more useful and effective.
  • the suggestion generator 230 takes a step further and ventures into a personal assist space. Instead of receiving user questions "can you recommend me of a show this weekend in San Francisco? can you book a flight to Seattle on Tuesday morning?” and answering the questions based on multiple applications, searches, comparisons and form fillings, the suggestion generator 230 asks questions for users and answers the questions on their behalf.
  • chat is becoming the go to application on user's smart phone, the suggestion generator 230 incorporating with other modules and engines now fits well into this area by drawing information from the chat and providing relevant updates.
  • the suggestion generator 230 communicates with other services to provide important updates in one place for all participants. Connecting users with other services by the suggestion generator 230 could also drive more users into other services since it wouldn't require users to download a separate mobile application for another service. Furthermore, the suggestion generator 230 offers developer components to build different type of services including understanding and translating user's conversations into actions, for example, voice commands and wear equipment.
  • the suggestion generator 230 comprises a response predictor 232 and a report module 234.
  • the response predictor 232 can be code and routines for detecting a question for a user in a conversation and determining a potential user reply to the question.
  • the response predictor 232 can be a set of instructions executable by the processor 202 to provide the functionality described below for detecting a question for a user in a conversation and determining a potential user reply to the question.
  • the response predictor 232 can be stored in the memory 204 of the computing device 200 and can be accessible and executable by the processor 202.
  • the response predictor 232 can be adapted for cooperation and communication with the processor 202 and other components of the computing device 200.
  • the response predictor 232 predicts users' potential answer and supplies users with a next reply, which is better than typing or copy-paste. For example, when a user is asked in a conversation when his or her flight is going to land, the response predictor 232 can make flight information available as a reply for the actionable entity "flight.”
  • the response predictor 232 predicts how a user would want to reply to a question based on the contextual indicators and the conversational analysis from the conversation analyzer 226 and the machine learning model from the model builder 224. For example, the response predictor 232 determines how a user tends to reply a certain type of questions based on what the machine learning model has learned from the conversation history and then offers one or more potential replies for the user. If the response predictor 232 learns that 30% of the time people respond A and 70% of the time people respond B to a question from the machine learning model, the response predictor 232 may supply answer B for a user that is going to answer the same question.
  • the response predictor 232 is at a particular advantage because of the contextual indicators and the conversational analysis received from the conversation analyzer 226. Knowing users' location, context, and other personal information enables the response predictor 232 to more easily and/or accurately predict a reply. For example, given the messages "I hate coffee” and "the tea house on the X street is the best" from user A in previous conversations, the response predictor 232 can easily supply the address of "the tea house on the X street" as the potential answer for user A when user B asks user A "which is your favorite place for coffee or tea?.”
  • the response predictor 232 also determines the type of answers based on the type of questions.
  • the response predictor 232 determines the type of questions based on the understanding of the sentence structure from the conversation analyzer 226. For example, the response predictor 232 provides A and B as potential answers if the question is "A or B?.” Or the response predictor 232 provides "yes” and “no” as potential answers when the question is "do you like A ?.”
  • the response predictor 232 also determines the type of questions based on other contextual indicators determined by the conversation analyzer 226.
  • the response predictor 232 provides a potential user reply in one-tap form and/or other single selection form. For example, assume the question is "can I send your contact info?.” The question includes an actionable entity "contact info.” Further assume a previous conversation includes "your" contact info. The response predictor 232 may provide a one-tap contact info as an answer.
  • a one-tap suggestion offered by the response predictor 232 based on the context of a conversation is convenient for a user in many ways (e.g., time, safety). For example, when user A sends user B a chat conversation, the response predictor 232 can send a one-tap suggestion saying that "I am driving" or even send "I will be home in 20 minutes" for user B to respond user A. When user B is flying and does not turn on the cell phone, the response predictor 232 can still automatically respond "I will be available four hours later" for user B based on a calendar.
  • An example interface showing one-tap replies is depicted in Figure 10.
  • the report module 234 can be code and routines for generating one or more reports related to a conversation based on organizing the conversation.
  • the report module 234 can be a set of instructions executable by the processor 202 to provide the functionality described below for generating one or more reports related to a conversation based on organizing the conversation.
  • the report module 234 can be stored in the memory 204 of the computing device 200 and can be accessible and executable by the processor 202.
  • the report module 234 can be adapted for cooperation and communication with the processor 202 and other components of the computing device 200.
  • the conversations may not be easily searchable, for example, a user may need to scroll up and down from a lot of search results to find a piece of information the user really wants.
  • the report module 234 aggregates information across conversations, organizes the conversations, and makes the conversations easily searchable by topics, objects, and/or other metadata.
  • the report module 234 receives conversations and contextual indicators determined from the conversations by the conversation analyzer 226, and organizes the conversations based on the contextual indicators. In some instances, the report module 234 organizes the conversations based on content. For example, the report module 234 organizes the conversations to find conversations that talk about a movie or mention a location. In other instances, the report module 234 organizes the conversations based on timeline. For example, the report module 234 organizes the conversations that mention a location in a descending time order. In other instances, the report module 234 organizes the conversations based on conversation flows.
  • the report module 234 retrieves conversation messages that include a location from the conversations, associates each message with a particular user who sent the message, and arranges the conversation messages in a descending time order. In another example, if a user says "click the search box," the report module 234 may determine that the next message in the conversation flow will be the word that the user specified in the search box and organizes the conversation in this order. In other instances, the report module 234 organizes the conversations based on sentiment and/or other feature(s). For example, the report module 234 organizes the conversations based on classifying people as electronic-book reader or paper-book reader. One skilled in the art, having had the benefit of the present disclosure, will understand that the report module 234 can organize the conversation based on other contextual indicators.
  • the report module 234 also labels the important message (e.g., an actionable entity) for later follow up.
  • the label can be flags, stars, etc.
  • the report module 234 can easily retrieve it.
  • the report module 234 obtains additional information by labelling the messages, for example, how often this message was mentioned, when a user mentioned this message, etc.
  • the report module 234 indexes the conversations (e.g., in a database stored in storage 210) based on organizing the conversations.
  • the report module 234 searches the conversations using the indexes, generates a report and returns the report to a user. For example, responsive to receiving a search request for conversations talking about a movie, the report module 234 searches conversation content about the movie.
  • the report module 234 organizes the search result based on ordering conversations according to the timeline of the conversations, associating each conversation message in a conversation with a user and a time, highlighting the important entities (e.g., movie names, theater names, show time, etc.,) and providing references back to the previous conversations based on the indexes.
  • the report module 234 generates a report for a user based on organizing the conversation.
  • An example conversation interface showing a conversation report is depicted in Figure 13.
  • the report module 234 also generates a summary by organizing and summarizing conversations within a time period. For example, the report module 234 generates a summary for a user to show how many conversations the user participated in within a week, who the user talks with in the week, what topics the user talked about in the week, how many photos the user shared in the week, etc.
  • An example conversation interface showing a conversation summary is depicted in Figure 14.
  • the report module 234 can retrieve the details for a user without necessitating that the user dig through and/or scroll through a verbatim transcript of a conversation.
  • the report module 234 also gets updates on topics discussed in the conversation for a user (e.g., game scores or shows).
  • the report module 234 provides not only a search result since users can look back at a meaningful trail of information shared based on events, places, people.
  • the report module 234 also allows users to interact with a collection of things they've shared or discussed such as a list of all restaurants they've talked about in the past few months.
  • the report module 234 provides a way to gain insights from conversations and to gauge context, sentiment, and/or other features. For example, the report module 234 can get valuable insights to gauge sentiment towards certain brands and to identify trending topics.
  • the user interface engine 236 can be code and routines for generating a conversation interface that displays conversation messages and suggestions of a conversation.
  • the user interface engine 236 can be a set of instructions executable by the processor 202 to provide the functionality described below for generating a user interface that displays conversation messages and suggestions of a conversation.
  • the user interface engine 236 can be stored in the memory 204 of the computing device 200 and can be accessible and executable by the processor 202.
  • the user interface engine 236 can be adapted for cooperation and communication with the processor 202 and other components of the computing device 200.
  • the conversation management module 222 instructs the user interface engine 236 to generate graphical data for displaying a conversation and
  • the suggestion generator 230 generates a suggestion for an actionable entity in a conversation.
  • the conversation management module 222 instructs the user interface engine 236 to generate graphical data for displaying the suggestion along with a conversation message in a conversation interface.
  • the user interface engine 236 generates multiple suggestion tools in a conversation interface for displaying one or more suggestions and receiving user interactions with the suggestion(s).
  • a suggestion includes information for a user to take an action on an entity.
  • the user interface engine 236 generates multiple suggestion tools for presenting the suggestion in parallel with the message including the entity.
  • the user interface engine 236 shows the map tool of ski resorts and a snow report along with the message "go skiing" on the same screen.
  • a user can view and select the map with the map tool.
  • the user interface engine 236 may show two calendar tools for two users along with the message discussing meeting up next day.
  • a user can view and schedule a calendar using the calendar tool.
  • the user interface engine 236 may show different purchase tools for dress and shoes along with the message discussing celebrity's outfit.
  • the user interface engine 236 may show a calendar tool highlighting times that two users are free next day and show a map tool highlighting coffee shops near addresses of both users along with the message "meet for coffee next day?" Or the user interface engine 236 may show a traffic map tool and a public transport tool along with the message "go from A to B?” In some instances, if a suggestion tool cannot fit into a screen (e.g., the small screen of a smart phone), the user interface engine 236 will show tool options first as multiple tools.
  • the user interface engine 236 generates multiple suggestion tools for serially presenting one or more suggestions. For example, upon receiving the message "meet up for a meal,” the user interface engine 236 shows options for lunch or dinner. Upon a user selection of lunch, the user interface engine 236 shows the restaurants open for lunch. Or upon receiving message "meet for coffee next day,” the user interface engine 236 shows a calendar tool first. The user interface engine 236 may also highlight times on the calendar when all users are available. Once a time is selected, the user interface engine 236 shows a map of coffee shops open at that time. Or upon receiving a message "buy skiing equipment in Los Angeles," the user interface engine 236 first shows a map of sports stores in Los Angeles. After a store is selected, the user interface engine 236 shows a purchase tool for equipment in that store. An example conversation interface showing serial presentation of suggestions is depicted in Figure 15.
  • Figure 3 includes a flowchart of an example method 300 for identifying an entity from a conversation and generating a suggestion for a user to take an action on the entity.
  • the method 300 begins by receiving 302 at least one conversation message from a conversation.
  • a conversation message may include a video message, an audio message, a text message, etc.
  • the method 300 then identifies 304 an entity that can be actionable from a conversation message and determines 306 contextual indicators of the entity.
  • the entity can be a person in a conversation message, an object in the conversation message, etc.
  • the conversation analyzer 226 discussed in further detail with reference to Figure 2 above may identify the entity and determine contextual indicators.
  • the conversation analyzer 226 may identify a location entity from a conversation message and determine contextual indicators of the location entity. For instance, for the location entity, the conversation analyzer 226 may determine contextual indicators based on when the location was mentioned, who mentioned the location in the conversation, which content of the conversation caused the location be perceived, which kind of content followed the location in the conversation, prior sentiment expressed toward the entity and/or related entities in prior conversations, etc.
  • the method 200 next determines 208 whether the entity is actionable based on the contextual indicators.
  • the determining engine 228 discussed in further detail with reference to Figure 2 may determine an actionable entity. For example, if the conversation analyzer 226 identifies the location entity "supermarket C" from a message "meet at supermarket C" between user A and user B in the conversation, the determining engine 228 may determine that this location entity "supermarket C" is actionable because users A and B will meet at the place. If the determining engine 228 determines that the entity is actionable, the method 300 moves to step 310. Otherwise, the method 300 ends.
  • the method 300 provides 310 a suggestion to a user to take an action on the entity.
  • the suggestion generator 230 discussed in further detail with reference to Figure 2 above may provide a suggestion. For example, the suggestion generator 230 provides a map to "supermarket C" and a public transport tool to direct how a user can get to the
  • the suggestion is provided to all participants of the conversation. In some other implementations, the suggestion is provided to a subset of the participants of the conversation. Moreover, in some implementations the suggestion may initially be provided to one or more participants without immediate incorporation of the suggestion into the transcript of the conversation. In some of those implementations, the suggestion may only be incorporated into the transcript in response to one or more (e.g., at least a threshold quantity of) providing "affirmative" user interface input directed toward the suggestion.
  • Figure 4 is a flowchart of a specific example method 400 for identifying an entity from a conversation and generating a suggestion for a user to take an action on the entity.
  • the method 400 receives 402 at least one conversation message from a conversation.
  • the method then performs 404 a natural language analysis (LP) on the at least one conversation message based on a machine learning model, and identifies 406 an entity that can be actionable from a conversation message based on the analysis.
  • LP natural language analysis
  • the conversation analyzer 226 determines entities "beautiful,” "weather” and "Store A" from a conversation message by parsing, extracting, and pattern matching.
  • the conversation analyzer 226 determines that "beautiful” is an adjective, "weather” is a noun, and "Store A” is a location based on the language semantic analysis using the SAFT library.
  • An entity identified by the conversation analyzer 226 can be actionable or non-actionable depending on whether it is determined that a user can take an action on it.
  • the method 400 determines 408 contextual indicators of the entity based on the analysis.
  • the conversation analyzer 226 performs the NLP based on pattern matching in combination with user information.
  • the conversation analyzer 226 can detect, from a conversation message, a "Coffee house” near user A instead of any arbitrary "Coffee house” in a city based on pattern matching in combination with user information of user A.
  • the conversation analyzer 226 may additionally or alternatively determine contextual indicators by performing sentiment analysis on a conversation message and associating the conversation message with a weight.
  • the conversation analyzer 226 receives messages "I hate coffee,” “tea is ok” and “coke is my favorite” from user A, and associates a most favorable to least favorable sentiment weight to coke, tea and coffee, respectively.
  • the weight indicates user A's beverage taste.
  • the conversation analyzer 226 may additionally and/or alternatively analyze a conversation to determine the tense (e.g., whether a conversation message is in the past tense) and/or the recency of the conversation message (e.g., whether a conversation message is new, how to weigh a new or old conversation message); and/or or collects user profile information (e.g., an age, a gender) and/or conversation metadata (e.g., a start time and an end time of a conversation, participants of the conversation, timestamps associated with each conversation message) as contextual indicators.
  • the tense e.g., whether a conversation message is in the past tense
  • the recency of the conversation message e.g., whether a conversation message is new, how to weigh a new or old conversation message
  • user profile information e.g., an age, a gender
  • conversation metadata e.g., a start time and an end time of a conversation, participants of the conversation, timestamps associated with each conversation message
  • the method 400 further tracks 410 conversation flows from each participant of the conversation and refines 412 the contextual indicators based on the tracking information.
  • the conversation analyzer 226 tracks users for possible participants of a conversation and determines which participant sends which message in what order. For example, the conversation analyzer 226 identifies an entity "hiking" from a message "do you want to go hiking?" sent from user A when talking with user B today.
  • the conversation analyzer 226 tracks conversation flows and determines that a message "I prefer fishing than hiking" from user B in a previous conversation should be used as contextual indicators for the entity "hiking.”
  • the method 400 determines 414 whether the entity is actionable based on the contextual indicators.
  • the determining engine 228 determines whether an entity is actionable based on a conversation flow. For example, user A enters a question "meet at ABC?" The determining engine 228 will not determine whether the location "ABC" is actionable until receiving user B's answer in the conversation flow.
  • the determining engine 228 determines whether an entity is actionable based on the tense of the message. For example, the determining engine 228 may determine an entity in a past tense message is non-actionable.
  • the determining engine 228 also determines whether an entity is actionable based on sentiment indicators and other contextual indicators.
  • the method 400 determines 416 a suggestion for a user to take an action on the entity based on the machine learning model and the contextual indicators. For example, if a user talks about taking a flight to Las Vegas next weekend, the determining engine 228 determines an actionable entity "flight" and the suggestion generator 230 offers a suggestion of a flight itinerary (e.g., time and airline) based on user' s previous purchase history learned in the machine learning model. The method 400 provides 418 the suggestion to the user to take an action on the entity.
  • a flight itinerary e.g., time and airline
  • Figure 5 is a flowchart 500 of an example method for manually determining an actionable entity from a conversation.
  • the method 500 receives 502 an indication to mark an entity in a conversation message as actionable.
  • the indication to mark the entity in a conversation message as actionable is received from a user (e.g., based on user interface input directed to term(s) in the conversation message that correspond to the entity).
  • the method 500 sends 504 the user indication, and optionally additional data (e.g., the conversation message, contextual indicators) as a training signal to a machine learning model.
  • the method 500 searches 506 for information relevant to the entity and provides 508 the information to the user.
  • a user determines "Green Park Museum” is the most important word in a message and indicates that this word is actionable by flagging the word. Responsive to receiving this indication, the determining engine 228 searches information related to "Green Park Museum” and retrieves the top five results to the user.
  • determining engine 228 also sends the indication as a training signal to train a machine learning model.
  • Figure 6 is a flowchart 600 of an example method for generating and updating a machine learning model.
  • the model builder 224 discussed in further detail with reference to Figure 2 above may generate and update a machine learning model.
  • the method 600 receives 602 data from a variety of sources including queries and document retrieval, extracts 604 features from the data and generates 606 a machine learning model based on the extracted features.
  • the model builder 224 includes a user' s search history into the model.
  • the model builder 224 upon receiving a collection of documents returned for a query and ranks associated with each document, the model builder 224 builds the model to include the ranks to show the relevancy between a document and the query.
  • the model builder 224 may also generate the model to include user statistics, user behavior pattern data, a sequence of weights and multipliers that are used to make predictions about a user action.
  • the model builder 224 generates the machine learning model and trains the model over time to understand the semantics of a conversation and derives meaning from the real time conversation. As the model evolves, the efficiency and accuracy of using the model in suggestion making, behavior prediction, etc., are improved.
  • the method 600 receive 608 a user action, the user action including a reaction to a suggestion provided for an entity, and trains 610 the machine learning model based on the user action. This may include generating a training example based on the user action and other associated data and training the machine learning model based on the training example.
  • the model builder 224 learns that a user never buys anything that was offered in a conversation interface with a purchase tool.
  • the model builder 224 receives a comment "I do not like this airline" regarding a flight itinerary suggestion. The model builder 224 takes this negative reaction as a signal to train the model. As a result, a different airline may be suggested to the user in the future.
  • FIG. 7 is a flowchart 700 of an example method for feeding entity information from a first application to a second application to generate a suggestion.
  • the method 700 receives 702 at least one conversation message from a conversation hosted by a first application and determines 704 that an entity in a conversation message is actionable.
  • the method 700 feeds 706 entity information to a second application, the entity information including participants of the conversation in the first application.
  • the method moves to step 708, where the second application generates a suggestion for a user to take an action on the entity. For example, if a flight number is mentioned in a conversation, the suggestion generator 230 notifies a third-party server to provide flight feeds (e.g., delay information) to a user.
  • flight feeds e.g., delay information
  • the suggestion generator 230 communicates with an advertisement application to use this information to target the user with advertising.
  • the method 700 provides 710 the suggestion to the participants of the conversation in the first application. For example, if user A mentions weather in a conversation with user B and user C, a weather service may send a hourly weather report to users A, B and C after receiving entity information of "weather" from the suggestion generator 230.
  • Figure 8 is a flowchart 800 of an example method for predicting a potential user reply in a conversation.
  • the method 800 receives 802 at least one conversation message from a conversation and determines 804 that an entity in a conversation message is actionable.
  • the method 800 detects 806 a question for a user in the conversation, the question being related to the entity and determines 808 a potential user reply to the question.
  • the response predictor 232 discussed in further detail with reference to Figure 2 below may detect a question and determine a potential user reply.
  • the response predictor 232 may supply answer B for a user that is going to answer the same question.
  • the response predictor 232 also determines a potential reply based on contextual indicators (e.g., users' location, context, and other personal information). For example, given the messages "I hate coffee” and "the tea house on the X street is the best" from user A in previous conversations, the response predictor 232 can easily supply the address of "the tea house on the X street" as the potential answer for user A when user B asks user A "which is your favorite place for coffee or tea?.”
  • the method 800 provides 810 the potential user reply in a one-tap form to the user.
  • the response predictor 232 can send a one-tap suggestion saying that "I am driving" or even send "I will be home in 20 minutes" for user B to respond user A.
  • the suggestion may then be incorporated in a transcript of the chat conversation.
  • the response predictor 232 can still automatically respond "I will be available four hours later" for user B based on a calendar.
  • Figure 9 is a flowchart 900 of an example method for selectively retrieving conversations and/or conversation messages in response to search requests.
  • the method 900 receives 902 at least one conversation and determines 904 contextual indicators from the at least one conversation.
  • the method 900 organizes 906 the at least one conversation based on the contextual indicators and indexes 908 the at least one conversation.
  • the report module 234 discussed in further detail with reference to Figure 2 above may organize and index the conversation.
  • the report module 234 may organize the conversations based on participant identities, entities referenced in the conversations, times of the conversation messages, other conversation content, and the like. In one example, the report module 234 may organize the conversations that mention a location in a descending time order.
  • the report module 234 may also organize the conversations based on conversation flows. For example, if a user says "I will search it," the report module 234 may determine that the next message in the conversation flow will be the word that the user specified in the search box and may organize the conversation in this order. The report module 234 may also organize and index the conversations based on sentiment and other contextual indicators.
  • the method 900 receives 910 a search request for the at least one conversation and retrieves 912 a portion of the at least one conversation based on indexing the at least one conversation. For example, responsive to receiving a search request for conversations talking about a movie, the report module 234 generates a report for a user based on organizing the conversations by searching conversation content about the movie, ordering conversations according to the timeline of the conversations, associating each conversation message in a conversation with a user and a time, highlighting the important entities (e.g., movie names, theater names, show time, etc.,) and providing references back to the previous conversations based on the indexes.
  • important entities e.g., movie names, theater names, show time, etc.
  • Figure 10 includes graphic representations of example conversation interfaces showing a suggestion.
  • the conversation interfaces 1000, 1050 are displayed on Aaron's mobile screen.
  • Aaron receives a message 1002 "Want to eat at Park Chow?" from Bob.
  • the suggestion application 132b determines that the location "Park Chow" is actionable. A suggestion therefore can be generated for Aaron and Bob to take an action on the location.
  • the suggestion application 132b generates a map 1004 of the location "Park Chow" for Aaron and/or Bob.
  • Aaron and/or Bob can view the map 1004 or share the map 1004 using the share button 1006.
  • the map 1004 is displayed twice before and after Aaron sends a message 1008 "Sure” to respond the message 1002.
  • the determining engine 228 determines that the location "Park Chow” is actionable only if Aaron positively answers the question 1002, for example, with the message 1008 "Sure".
  • the location suggestion (e.g., the map 1004) may only be displayed once after Aaron sends the message 1008. In other words, the location suggestion may not be incorporated into the transcript of the conversation until Aaron sends the message 1008 via interaction with the affirmative interaction element "Sure"".
  • Aaron and Bob may
  • Figure 11 A is a graphic representation of an example conversation interface showing a conversation generated by a conversation service.
  • An ongoing conversation between users Melissa G., Alice Z., Chris B. is displayed on the conversation interface 1100.
  • the users Melissa G., Alice Z., Chris B may be members of a social network in this non- limiting example.
  • the conversation session between the users is generated and managed by a first application (e.g., a suggestion application 132) in the social network.
  • the conversation interface 1 100 can be displayed on Melissa's computer screen.
  • the users converse about a plan that night, Melissa expresses interest in shopping and states that "Dress Factory is good.
  • the suggestion application 132b identifies that the important entities in this message are "Dress Factory” and "discount," the suggestion application 132b generates a suggestion based on these important entities. However, instead of generating a suggestion by the suggestion application 132b as in Figure 10, the suggestion application 132b feeds the information "Dress Factory” and "discount” to another service (e.g., a third-party promotion service) to generate a suggestion.
  • another service e.g., a third-party promotion service
  • Figure 1 IB is a graphic representation of an example conversation interface showing a suggestion generated by a third-party promotion service.
  • the conversation interface 1 150 is displayed on Melissa's mobile screen.
  • the conversation interface 1 150 includes a suggestion 1 152 "Dress Factory has 20% off for dress and jewelry until May 30.”
  • the promotion service generates suggestion 1 152 based on the entities "Dress Factory” and "discount” identified by the suggestion application 132b in Figure 1 1 A (e.g., after being fed that information by a separate server handling the conversation).
  • the promotion service may perform a search for the entities on the Internet to generate the suggestion 1 152.
  • the suggestion generator 132b may receive the suggestion 1 152 and then provide the suggestion 1 152 to all, or a subset of, participants of the conversation in Figure 1 1 A, e.g., Melissa G., Alice Z., and/or Chris B.
  • Figure 1 IB displays the suggestion 1 152 on a mobile screen of Melissa, identical to the suggestions shown to Alice or Chris.
  • Figure 12 is a graphic representation of an example conversation interface showing one-tap replies.
  • the conversation interface 1200 is displayed on Aaron's mobile screen.
  • the suggestion application 132b may predict a potential response from Aaron based on previous user actions of Aaron (e.g., a purchase history), and generates a one-tap reply 1204 for Aaron.
  • the auto reply 1004 lists the flight number, the arrival time and a flight status. Aaron therefore saves time to type an answer to Bob.
  • the suggestion application 132b determines that this is a yes or no question, and generates a one-tap auto reply 1008 to include two options "YES” or "NO” for Aaron to choose.
  • Figure 13 is a graphic representation of an example conversation interface showing a conversation report.
  • the conversation interface 1300 includes a search box 1302, where the user can search for conversations based on participant name, topic, time, combinations thereof, etc.
  • the user searched for conversations about a movie "star wars.”
  • the suggestion application 132b may generate a report.
  • the report may start with a summary 1304 on when the user discussed the topic and the other participants in the conversations, e.g., that Mellissa, Alice and Jenny discussed "star wars" between 7: 10- 7:30pm on Wednesday.
  • the report may also include the conversation messages from each participant in a time order.
  • the report may also highlight the keyword "star wars" 1306 searched by the user.
  • the report may highlight certain entities (e.g., actors “AA” 1108 and "BB” 1310), to enable the user to retrieve more information (e.g., biography, photos) on the entities.
  • Each highlighted and identified entity may be associated with a hyperlink that enables the user to perform a search for the entity on the Internet.
  • Figure 14 is a graphic representation of an example conversation interface showing a conversation summary.
  • the conversation interface 1400 includes a conversation summary 1402 generated by the suggestion application 132b for a user summarizing the conversations that the user participated in during a week.
  • the conversation summary may include two exemplary portions 1402 and 1404.
  • the summary may include the number of conversations the user participated in during the past week, and some basic information on the conversations, e.g., time of day, date, identities of the other participants, and the like.
  • the first portion 1402 may also include a detail option 1406 for each conversation that provides further details on each conversation including conversation messages from each participant, the time when each message was sent, etc.
  • the second portion 1404 may include salient information on the conversations conducted in the past week including, for example, the topics discussed in the conversations (optionally, a pie chart 1408 describing the percentage of each topic in the conversations), audio/visual content shared in the conversations (optionally, a link 1410 to play the content), the photos 1412 shared in the two conversations.
  • Figure 15 includes graphic representations of example conversation interfaces showing serial presentation of suggestions.
  • the conversation interface 1500 is displayed on a user's mobile screen. Responsive to receiving a message 1502 "meet up for a meal?," the suggestion application 132b generates and displays options 1504 for "lunch,” “dinner,” “brunch,” etc. The user may select one of the options 1504 first suggested to the user and triggers a second suggestion to be generated.
  • Figure 15 displays a conversation interface 1550 generated responsive to receiving the user selection as shown in the option 1506.
  • the suggestion application 132b Upon the user's selection for lunch in the option 1506, the suggestion application 132b generates and displays certain restaurants open for lunch. In certain implementations, the suggestion application 132b may display only those restaurants that are within a certain distance range of all of the participants in the conversation.
  • the restaurant information 1508 may include a restaurant name, a restaurant type, a review and a price.
  • the displayed restaurant information may be selectable and actionable.
  • the user may select a particular restaurant and display it to the other participants in the conversation, e.g., as a message including the restaurant name, as a message including the restaurant name and prefilled default words suitable to the conversation (e.g., "let's meet at... ").
  • the user may select a particular restaurant to perform a search for the restaurant on the Internet, retrieve merchant information from a website, retrieve a map of the restaurant location, and the like. The user may then choose to display the retrieved information to the other participants in the conversation, e.g., by inserting a map of the restaurant location into the conversation interface.
  • the specification also relates to an apparatus for performing the operations herein.
  • This apparatus may be specially constructed for the required purposes, or it may include a general-purpose computer selectively activated or reconfigured by a computer program stored in the computer.
  • a computer program may be stored in a non-transitory computer readable storage medium, such as, but is not limited to, any type of disk including floppy disks, optical disks, CD-ROMs, and magnetic disks, read-only memories (ROMs), random access memories (RAMs), erasable programmable read-only memory (EPROMs), electrically erasable programmable read-only memory (EEPROMs), magnetic or optical cards, flash memories including USB keys with non-volatile memory or any type of media suitable for storing electronic instructions, each coupled to a computer system bus.
  • ROMs read-only memories
  • RAMs random access memories
  • EPROMs erasable programmable read-only memory
  • EEPROMs electrically erasable programmable read-only memory
  • the specification can take the form of an entirely hardware embodiment, an entirely software embodiment or embodiments containing both hardware and software elements.
  • the specification is implemented in software, which includes but is not limited to firmware, resident software, microcode, etc.
  • the description can take the form of a computer program product accessible from a non-transitory computer-usable or computer-readable medium providing program code for use by or in connection with a computer or any instruction execution system.
  • a non-transitory computer-usable or computer readable medium can be any apparatus that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.
  • a data processing system suitable for storing and/or executing program code will include at least one processor coupled directly or indirectly to memory elements through a system bus.
  • the memory elements can include local memory employed during actual execution of the program code, bulk storage, and cache memories which provide temporary storage of at least some program code in order to reduce the number of times code must be retrieved from bulk storage during execution.
  • I/O devices including but not limited to keyboards, displays, pointing devices, etc.
  • I/O controllers can be coupled to the system either directly or through intervening I/O controllers.
  • Network adapters may also be coupled to the system to enable the data processing system to become coupled to other data processing systems or remote printers or social network data stores through intervening private or public networks.
  • Modems, cable modem and Ethernet cards are just a few of the currently available types of network adapters.
  • the users may be provided with an opportunity to control whether programs or features collect user information (e.g., information about a user's social network, social actions or activities, profession, a user's preferences, or a user's current location), or to control whether and/or how to receive content from the content server that may be more relevant to the user.
  • user information e.g., information about a user's social network, social actions or activities, profession, a user's preferences, or a user's current location
  • certain data may be treated in one or more ways before it is stored or used, so that personally identifiable information is removed.
  • a user's identity may be treated so that no personally identifiable information can be determined for the user, or a user's geographic location may be generalized where location information is obtained (such as to a city, ZIP code, or state level), so that a particular location of a user cannot be determined.
  • location information such as to a city, ZIP code, or state level
  • the user may have control over how information is collected about the user and used by the various components of the system.
  • modules, routines, features, attributes, methodologies and other aspects of the disclosure can be implemented as software, hardware, firmware or any combination of the three.
  • a component, an example of which is a module, of the specification is implemented as software
  • the component can be implemented as a standalone program, as part of a larger program, as a plurality of separate programs, as a statically or dynamically linked library, as a kernel loadable module, as a device driver, and/or in every and any other way known now or in the future to those of ordinary skill in the art of computer programming.
  • the disclosure is in no way limited to implementation in any specific programming language, or for any specific operating system or environment. Accordingly, the disclosure is intended to be illustrative, but not limiting, of the scope of the specification, which is set forth in the following claims.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • User Interface Of Digital Computer (AREA)
  • Information Transfer Between Computers (AREA)
  • Machine Translation (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
PCT/US2016/068059 2015-12-21 2016-12-21 Automatic suggestions for message exchange threads Ceased WO2017112786A1 (en)

Priority Applications (4)

Application Number Priority Date Filing Date Title
EP16825663.4A EP3395018A1 (en) 2015-12-21 2016-12-21 Automatic suggestions for message exchange threads
JP2018551908A JP6638087B2 (ja) 2015-12-21 2016-12-21 メッセージ交換スレッド用の自動提案
CN201680082643.2A CN108781175B (zh) 2015-12-21 2016-12-21 用于消息交换题绪的自动建议的方法、介质及系统
KR1020187019756A KR20180093040A (ko) 2015-12-21 2016-12-21 메시지 교환 스레드들에 대한 자동적인 제안들

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US201562270454P 2015-12-21 2015-12-21
US62/270,454 2015-12-21
US15/386,162 US10530723B2 (en) 2015-12-21 2016-12-21 Automatic suggestions for message exchange threads
US15/386,162 2016-12-21

Publications (1)

Publication Number Publication Date
WO2017112786A1 true WO2017112786A1 (en) 2017-06-29

Family

ID=59066635

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/US2016/068059 Ceased WO2017112786A1 (en) 2015-12-21 2016-12-21 Automatic suggestions for message exchange threads

Country Status (6)

Country Link
US (2) US10530723B2 (enExample)
EP (1) EP3395018A1 (enExample)
JP (1) JP6638087B2 (enExample)
KR (1) KR20180093040A (enExample)
CN (1) CN108781175B (enExample)
WO (1) WO2017112786A1 (enExample)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102016125509A1 (de) * 2016-05-17 2017-11-23 Google Inc. Einbindung von auswählbaren Anwendungsverknüpfungen in Nachrichtenaustausch-Threads
US10291565B2 (en) 2016-05-17 2019-05-14 Google Llc Incorporating selectable application links into conversations with personal assistant modules
US11176520B2 (en) 2019-04-18 2021-11-16 Microsoft Technology Licensing, Llc Email content modification system

Families Citing this family (124)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
SG190612A1 (en) 2009-06-09 2013-06-28 Cimas Ltd Halogenated aliphatic carboxylic acids, oligomers and/or polymers thereof and their use in devitalizing external and internal neoplasms
US10430894B2 (en) 2013-03-21 2019-10-01 Khoros, Llc Gamification for online social communities
US9213941B2 (en) 2014-04-22 2015-12-15 Google Inc. Automatic actions based on contextual replies
WO2017112786A1 (en) 2015-12-21 2017-06-29 Google Inc. Automatic suggestions for message exchange threads
CN108476164B (zh) 2015-12-21 2021-10-08 谷歌有限责任公司 在消息传送应用中自动地提供机器人服务的方法
CN106910513A (zh) * 2015-12-22 2017-06-30 微软技术许可有限责任公司 情绪智能聊天引擎
US10846475B2 (en) * 2015-12-23 2020-11-24 Beijing Xinmei Hutong Technology Co., Ltd. Emoji input method and device thereof
US10212119B2 (en) * 2016-01-13 2019-02-19 Pooja Bisarya Method and system for enabling messaging between users
US20170222956A1 (en) * 2016-02-01 2017-08-03 Priya Jain Method and system for enabling an auto response facility between users
US10757048B2 (en) 2016-04-08 2020-08-25 Microsoft Technology Licensing, Llc Intelligent personal assistant as a contact
US10158593B2 (en) * 2016-04-08 2018-12-18 Microsoft Technology Licensing, Llc Proactive intelligent personal assistant
US10904168B2 (en) * 2016-06-03 2021-01-26 Spotify Ab System and method for providing digital media content with a conversational messaging environment
US10387461B2 (en) 2016-08-16 2019-08-20 Google Llc Techniques for suggesting electronic messages based on user activity and other context
CN109952572B (zh) 2016-09-20 2023-11-24 谷歌有限责任公司 基于消息贴纸的建议响应
DE112017003594T5 (de) 2016-09-20 2019-04-25 Google Llc Bot, der Genehmigung zum Zugriff auf Daten anfordert
US10015124B2 (en) * 2016-09-20 2018-07-03 Google Llc Automatic response suggestions based on images received in messaging applications
US10846618B2 (en) 2016-09-23 2020-11-24 Google Llc Smart replies using an on-device model
US11159473B2 (en) * 2016-10-10 2021-10-26 Facebook, Inc. Systems and methods for sharing content
US10416846B2 (en) 2016-11-12 2019-09-17 Google Llc Determining graphical element(s) for inclusion in an electronic communication
CN108075959B (zh) * 2016-11-14 2021-03-12 腾讯科技（深圳）有限公司 一种会话消息处理方法和装置
US10146768B2 (en) 2017-01-25 2018-12-04 Google Llc Automatic suggested responses to images received in messages using language model
US10614164B2 (en) * 2017-02-27 2020-04-07 International Business Machines Corporation Message sentiment based alert
US11494440B1 (en) * 2017-04-12 2022-11-08 Meta Platforms, Inc. Proactive and reactive suggestions for a messaging system
US10902462B2 (en) 2017-04-28 2021-01-26 Khoros, Llc System and method of providing a platform for managing data content campaign on social networks
US10044862B1 (en) * 2017-04-28 2018-08-07 International Business Machines Corporation Dynamic topic guidance in the context of multi-round conversation
US10891485B2 (en) 2017-05-16 2021-01-12 Google Llc Image archival based on image categories
US10348658B2 (en) 2017-06-15 2019-07-09 Google Llc Suggested items for use with embedded applications in chat conversations
US10404636B2 (en) 2017-06-15 2019-09-03 Google Llc Embedded programs and interfaces for chat conversations
US10917483B2 (en) * 2017-06-22 2021-02-09 Numberai, Inc. Automated communication-based intelligence engine
FR3067554A1 (fr) * 2017-06-23 2018-12-14 Orange Procede et dispositif de traitement de messages textuels recus par une application de messagerie executee dans un terminal de communication d'un utilisateur
US11699039B2 (en) * 2017-06-28 2023-07-11 Microsoft Technology Licensing, Llc Virtual assistant providing enhanced communication session services
US10585991B2 (en) * 2017-06-29 2020-03-10 Microsoft Technology Licensing, Llc Virtual assistant for generating personalized responses within a communication session
US11258734B1 (en) 2017-08-04 2022-02-22 Grammarly, Inc. Artificial intelligence communication assistance for editing utilizing communication profiles
US20190050774A1 (en) * 2017-08-08 2019-02-14 General Electric Company Methods and apparatus to enhance emotional intelligence using digital technology
US10951558B2 (en) * 2017-09-27 2021-03-16 Slack Technologies, Inc. Validating application dialog associated with a triggering event identification within user interaction data received via a group-based communication interface
US10999278B2 (en) 2018-10-11 2021-05-04 Spredfast, Inc. Proxied multi-factor authentication using credential and authentication management in scalable data networks
US10346449B2 (en) 2017-10-12 2019-07-09 Spredfast, Inc. Predicting performance of content and electronic messages among a system of networked computing devices
US11570128B2 (en) 2017-10-12 2023-01-31 Spredfast, Inc. Optimizing effectiveness of content in electronic messages among a system of networked computing device
US11470161B2 (en) 2018-10-11 2022-10-11 Spredfast, Inc. Native activity tracking using credential and authentication management in scalable data networks
US10785222B2 (en) 2018-10-11 2020-09-22 Spredfast, Inc. Credential and authentication management in scalable data networks
US11050704B2 (en) 2017-10-12 2021-06-29 Spredfast, Inc. Computerized tools to enhance speed and propagation of content in electronic messages among a system of networked computing devices
US10652189B2 (en) 2017-10-19 2020-05-12 Chicago Mercantile Exchange Inc. Message encoding and transmission across multiple platforms
US11113604B2 (en) * 2017-11-06 2021-09-07 Google Llc Training and/or utilizing an interaction prediction model to determine when to interact, and/or prompt for interaction, with an application on the basis of an electronic communication
US10601937B2 (en) * 2017-11-22 2020-03-24 Spredfast, Inc. Responsive action prediction based on electronic messages among a system of networked computing devices
US10609511B2 (en) * 2017-12-08 2020-03-31 Facebook, Inc. Techniques for providing meeting location suggestions
US10891526B2 (en) 2017-12-22 2021-01-12 Google Llc Functional image archiving
US10587553B1 (en) * 2017-12-29 2020-03-10 Entefy Inc. Methods and systems to support adaptive multi-participant thread monitoring
US11061900B2 (en) 2018-01-22 2021-07-13 Spredfast, Inc. Temporal optimization of data operations using distributed search and server management
US10594773B2 (en) 2018-01-22 2020-03-17 Spredfast, Inc. Temporal optimization of data operations using distributed search and server management
JP7073122B2 (ja) * 2018-01-31 2022-05-23 Ｄｙｎａｂｏｏｋ株式会社 電子機器、制御方法及びプログラム
US10866989B1 (en) * 2018-04-06 2020-12-15 Architecture Technology Corporation Real time recommender modeling system, methods of construction, and methods of use
US20190327200A1 (en) * 2018-04-20 2019-10-24 Microsoft Technology Licensing, Llc Customized controls for handling communication content
US11322264B2 (en) 2018-04-23 2022-05-03 DNAFeed Inc. Systems and methods for human-augmented communications
US11074284B2 (en) * 2018-05-07 2021-07-27 International Business Machines Corporation Cognitive summarization and retrieval of archived communications
US20190384811A1 (en) * 2018-06-14 2019-12-19 Pubali Sen System and method for communication exchange feedback
US10901577B2 (en) 2018-07-17 2021-01-26 Google Llc Methods and systems for input suggestion
US10721190B2 (en) * 2018-07-31 2020-07-21 Microsoft Technology Licensing, Llc Sequence to sequence to classification model for generating recommended messages
US11252117B2 (en) * 2018-08-10 2022-02-15 International Business Machines Corporation Electronic communication message coaching service
CN115550304B (zh) * 2018-08-22 2023-10-10 谷歌有限责任公司 确定一组用户的活动实例集合的方法、设备和存储介质
US10909327B2 (en) * 2018-08-24 2021-02-02 International Business Machines Corporation Unsupervised learning of interpretable conversation models from conversation logs
US11182860B2 (en) * 2018-10-05 2021-11-23 The Toronto-Dominion Bank System and method for providing photo-based estimation
US10855657B2 (en) 2018-10-11 2020-12-01 Spredfast, Inc. Multiplexed data exchange portal interface in scalable data networks
US11238226B2 (en) * 2018-11-15 2022-02-01 Nuance Communications, Inc. System and method for accelerating user agent chats
US11048767B2 (en) * 2018-11-16 2021-06-29 Sap Se Combination content search
US12411843B2 (en) * 2018-11-16 2025-09-09 International Business Machines Corporation Intelligent resolution of conflicting information in adaptive dialog systems
US11777874B1 (en) * 2018-12-14 2023-10-03 Carvana, LLC Artificial intelligence conversation engine
CN110209952B (zh) * 2018-12-18 2023-03-24 腾讯科技（深圳）有限公司 信息推荐方法、装置、设备及存储介质
US10956474B2 (en) 2019-03-14 2021-03-23 Microsoft Technology Licensing, Llc Determination of best set of suggested responses
EP3942510A4 (en) * 2019-03-20 2022-04-27 Samsung Electronics Co., Ltd. METHOD AND SYSTEM FOR DELIVERING PERSONALIZED MULTIMODAL OBJECTS IN REAL TIME
JP7104277B2 (ja) * 2019-03-29 2022-07-21 株式会社Aill コミュニケーション支援サーバ、コミュニケーション支援システム、コミュニケーション支援方法、及びコミュニケーション支援プログラム
WO2020202315A1 (ja) * 2019-03-29 2020-10-08 株式会社Ａｉｌｌ コミュニケーション支援サーバ、コミュニケーション支援システム、コミュニケーション支援方法、及びコミュニケーション支援プログラム
JP7409781B2 (ja) * 2019-04-01 2024-01-09 Ｌｉｎｅヤフー株式会社 出力プログラム、出力装置及び出力方法
US10931540B2 (en) 2019-05-15 2021-02-23 Khoros, Llc Continuous data sensing of functional states of networked computing devices to determine efficiency metrics for servicing electronic messages asynchronously
WO2020246020A1 (ja) * 2019-06-07 2020-12-10 富士通株式会社 情報管理プログラム、情報管理方法および情報管理装置
US10902190B1 (en) * 2019-07-03 2021-01-26 Microsoft Technology Licensing Llc Populating electronic messages with quotes
US12405705B2 (en) * 2019-07-11 2025-09-02 Google Llc System and method for providing an artificial intelligence control surface for a user of a computing device
CN112448879B (zh) * 2019-08-27 2022-09-13 腾讯科技（深圳）有限公司 提醒消息推送方法、装置、终端及计算机可读存储介质
US11551242B2 (en) * 2019-12-13 2023-01-10 Jpmorgan Chase Bank, N.A. System and method for providing intelligent dashboards for critical business flows
US11521611B2 (en) 2019-12-19 2022-12-06 Palo Alto Research Center Incorporated Using conversation structure and content to answer questions in multi-part online interactions
US11555622B2 (en) * 2019-12-19 2023-01-17 International Business Machines Corporation Intelligent context-based control of air flow
NL2024677B1 (en) * 2020-01-15 2021-09-08 Microsoft Technology Licensing Llc Generating suggestions for users based on identifying direct interactions in group chats
US12143527B2 (en) * 2020-02-20 2024-11-12 Grey Wall Software Llc Computer messaging analytics systems and methods
US11023298B1 (en) * 2020-02-28 2021-06-01 Snap Inc. Displaying items based on a configuration file
US11462220B2 (en) * 2020-03-04 2022-10-04 Accenture Global Solutions Limited Infrastructure automation platform to assist in performing actions in response to tasks
CA3171846A1 (en) * 2020-04-15 2021-10-21 Benjamin Lloyd Styles Systems and computer-implemented methods for capital management
US10951564B1 (en) 2020-04-17 2021-03-16 Slack Technologies, Inc. Direct messaging instance generation
US11436408B2 (en) * 2020-07-01 2022-09-06 Motorola Mobility Llc Targeted natural language responses for electronic communications
US11438289B2 (en) 2020-09-18 2022-09-06 Khoros, Llc Gesture-based community moderation
US12120078B2 (en) 2020-09-18 2024-10-15 Khoros, Llc Automated disposition of a community of electronic messages under moderation using a gesture-based computerized tool
US11128589B1 (en) 2020-09-18 2021-09-21 Khoros, Llc Gesture-based community moderation
US11784949B2 (en) 2020-10-06 2023-10-10 Salesforce, Inc. Limited functionality interface for communication platform
US11741949B2 (en) 2020-10-30 2023-08-29 Nvidia Corporation Real-time video conference chat filtering using machine learning models
US11283751B1 (en) * 2020-11-03 2022-03-22 International Business Machines Corporation Using speech and facial bio-metrics to deliver text messages at the appropriate time
US11438282B2 (en) 2020-11-06 2022-09-06 Khoros, Llc Synchronicity of electronic messages via a transferred secure messaging channel among a system of various networked computing devices
US12197875B2 (en) 2021-07-31 2025-01-14 Khoros, Llc Automated predictive response computing platform implementing adaptive data flow sets to exchange data via an omnichannel electronic communication channel independent of data source
US11627100B1 (en) 2021-10-27 2023-04-11 Khoros, Llc Automated response engine implementing a universal data space based on communication interactions via an omnichannel electronic data channel
US12158903B2 (en) 2020-11-06 2024-12-03 Khoros, Llc Automated response engine to implement internal communication interaction data via a secured omnichannel electronic data channel and external communication interaction data
US11924375B2 (en) 2021-10-27 2024-03-05 Khoros, Llc Automated response engine and flow configured to exchange responsive communication data via an omnichannel electronic communication channel independent of data source
US11228644B1 (en) * 2020-11-10 2022-01-18 Capital One Services, Llc Systems and methods to generate contextual threads
US11714629B2 (en) 2020-11-19 2023-08-01 Khoros, Llc Software dependency management
KR20220071487A (ko) * 2020-11-24 2022-05-31 라인 가부시키가이샤 대화방 내에서 검색 기능을 제공하는 방법, 시스템, 및 컴퓨터 프로그램
CN112492355B (zh) 2020-11-25 2022-07-08 北京字跳网络技术有限公司 发布和回复多媒体内容的方法、装置和设备
US11792141B2 (en) * 2021-01-12 2023-10-17 Meta Platforms Technologies, Llc Automated messaging reply-to
US12014731B2 (en) 2021-01-29 2024-06-18 Zoom Video Communications, Inc. Suggesting user actions during a video conference
JP7677816B2 (ja) * 2021-03-26 2025-05-15 本田技研工業株式会社 情報提供システム、及び情報提供システムの制御方法
US11816097B2 (en) 2021-06-08 2023-11-14 Capital One Services, Llc Geographical location determination system
US11743215B1 (en) 2021-06-28 2023-08-29 Meta Platforms Technologies, Llc Artificial reality messaging with destination selection
US20230050045A1 (en) * 2021-08-12 2023-02-16 Yohana Llc Systems and methods for proposal acceptance in a task determination system
AU2022337282A1 (en) * 2021-09-01 2024-03-14 Panasonic Well Llc Systems and methods for generating and presenting dynamic task summaries
US12332934B2 (en) 2023-04-11 2025-06-17 Khoros, Llc Automated response engine implementing a universal data space based on communication interactions via an omnichannel electronic data channel
US11822562B2 (en) 2022-03-31 2023-11-21 Capital One Services, Llc Unstructured text processing for geographical location determination system
US12235889B2 (en) 2022-08-26 2025-02-25 Google Llc Device messages provided in displayed image compilations based on user content
AT526801A1 (de) * 2022-12-15 2024-07-15 Frequentis Ag Kommunikationssystem für textbasierte Kommunikation
US12261844B2 (en) 2023-03-06 2025-03-25 Spredfast, Inc. Multiplexed data exchange portal interface in scalable data networks
US12348465B2 (en) 2023-04-25 2025-07-01 Zoom Communications, Inc. Facilitating user actions for virtual interactions
USD1054441S1 (en) * 2023-05-01 2024-12-17 Airbnb, Inc. Display screen with graphical user interface
US12293335B2 (en) * 2023-05-04 2025-05-06 Christina Massa System and method of automating the recruitment process between employment candidates and recruiting entities
US20240380714A1 (en) * 2023-05-11 2024-11-14 Douglas VORT Systems and methods for visual cue based messaging
TWI846528B (zh) * 2023-06-29 2024-06-21 英業達股份有限公司 具主動聊天應答的客製化設定及更新下載系統及其方法
US20250023930A1 (en) * 2023-07-14 2025-01-16 Google Llc Artificial intelligence (ai)-based document retrieval during a virtual meeting
US12149497B1 (en) * 2023-07-26 2024-11-19 AsqMe Inc. Automation of online response to online query
US20250039334A1 (en) * 2023-07-28 2025-01-30 Zoom Video Communications, Inc. Collaboration using conversational artificial intelligence during video conferencing
JP2025044256A (ja) * 2023-09-19 2025-04-01 ソフトバンクグループ株式会社 システム
US12348476B2 (en) * 2023-10-30 2025-07-01 Zoom Communications, Inc. Message generation based on multichannel context

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140237057A1 (en) * 2013-02-21 2014-08-21 Genesys Telecommunications Laboratories, Inc. System and method for processing private messages in a contact center
US20150178388A1 (en) * 2013-12-19 2015-06-25 Adobe Systems Incorporated Interactive communication augmented with contextual information
US20150302301A1 (en) * 2014-04-22 2015-10-22 Google Inc. Automatic actions based on contextual replies

Family Cites Families (306)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5963649A (en) 1995-12-19 1999-10-05 Nec Corporation Message authorization system for authorizing message for electronic document
US6092102A (en) 1997-10-24 2000-07-18 University Of Pittsburgh Of The Commonwealth System Of Higher Education System and method for notifying users about information or events of an enterprise
US7904187B2 (en) 1999-02-01 2011-03-08 Hoffberg Steven M Internet appliance system and method
JP2000298676A (ja) 1999-04-14 2000-10-24 Bandai Co Ltd 情報提供装置
US7216080B2 (en) 2000-09-29 2007-05-08 Mindfabric Holdings Llc Natural-language voice-activated personal assistant
JP2002132804A (ja) 2000-10-24 2002-05-10 Sanyo Electric Co Ltd ユーザ支援システム
US20020103837A1 (en) 2001-01-31 2002-08-01 International Business Machines Corporation Method for handling requests for information in a natural language understanding system
WO2003024094A1 (en) 2001-09-10 2003-03-20 Nikon Corporation Digital camera system, image storage apparatus, and digital camera
US7636750B2 (en) 2001-10-24 2009-12-22 Sprint Spectrum L.P. Method and system for controlling scope of user participation in a communication session
US7283992B2 (en) 2001-11-30 2007-10-16 Microsoft Corporation Media agent to suggest contextually related media content
US20040001099A1 (en) 2002-06-27 2004-01-01 Microsoft Corporation Method and system for associating actions with semantic labels in electronic documents
US7234117B2 (en) 2002-08-28 2007-06-19 Microsoft Corporation System and method for shared integrated online social interaction
US20110107223A1 (en) 2003-01-06 2011-05-05 Eric Tilton User Interface For Presenting Presentations
WO2004077291A1 (ja) 2003-02-25 2004-09-10 Matsushita Electric Industrial Co., Ltd. アプリケーションプログラムの予測方法及び移動体端末
JP4647610B2 (ja) 2003-05-16 2011-03-09 グーグル インコーポレイテッド ネットワーク接続されたチャット及びメディア共有のシステム及び方法
US8645697B1 (en) 2003-08-08 2014-02-04 Radix Holdings, Llc Message authorization
US7363029B2 (en) 2003-09-12 2008-04-22 Core Mobility, Inc. Unified interface for voice, text or picture message authoring
US20120322428A1 (en) 2004-09-30 2012-12-20 Motedata Inc. Network of tags
US10635723B2 (en) 2004-02-15 2020-04-28 Google Llc Search engines and systems with handheld document data capture devices
US7363295B2 (en) 2004-04-19 2008-04-22 Yahoo! Inc. Techniques for inline searching in an instant messenger environment
WO2005125070A2 (en) 2004-06-14 2005-12-29 Semandex Networks, Inc. System and method for providing content-based instant messaging
US7464110B2 (en) 2004-06-30 2008-12-09 Nokia Corporation Automated grouping of image and other user data
US7734927B2 (en) 2004-07-21 2010-06-08 International Business Machines Corporation Real-time voting based authorization in an autonomic workflow process using an electronic messaging system
US7917582B2 (en) 2004-07-27 2011-03-29 Siemens Enterprise Communications, Inc. Method and apparatus for autocorrelation of instant messages
US7720436B2 (en) 2006-01-09 2010-05-18 Nokia Corporation Displaying network objects in mobile devices based on geolocation
US20060150119A1 (en) 2004-12-31 2006-07-06 France Telecom Method for interacting with automated information agents using conversational queries
US7197324B2 (en) 2005-01-31 2007-03-27 Sweeney Robert J Permission based text messaging
US7603413B1 (en) 2005-04-07 2009-10-13 Aol Llc Using automated agents to facilitate chat communications
US7860319B2 (en) 2005-05-11 2010-12-28 Hewlett-Packard Development Company, L.P. Image management
US7512580B2 (en) 2005-08-04 2009-03-31 Sap Ag Confidence indicators for automated suggestions
CN1988461A (zh) 2005-12-23 2007-06-27 腾讯科技（深圳）有限公司 一种即时通信工具的聊天场景音乐播放方法和系统
US7747785B2 (en) 2006-04-14 2010-06-29 Microsoft Corporation Instant messaging plug-ins
US9318108B2 (en) 2010-01-18 2016-04-19 Apple Inc. Intelligent automated assistant
WO2008045811A2 (en) 2006-10-10 2008-04-17 Orgoo, Inc. Integrated electronic mail and instant messaging system
US20080120371A1 (en) 2006-11-16 2008-05-22 Rajat Gopal Relational framework for non-real-time audio/video collaboration
US8671021B2 (en) 2006-12-13 2014-03-11 Quickplay Media Inc. Consumption profile for mobile media
JP2008191748A (ja) 2007-02-01 2008-08-21 Oki Electric Ind Co Ltd ユーザ間コミュニケーション方法、ユーザ間コミュニケーションプログラム、ユーザ間コミュニケーション装置
US8654176B2 (en) 2007-06-14 2014-02-18 Sharp Kabushiki Kaisha Operating system that includes an image data receiving device and an operation device for processing image data sets
JP2009009334A (ja) 2007-06-27 2009-01-15 Ricoh Co Ltd 画像処理装置、画像処理方法及び画像処理プログラム
CN101159576B (zh) 2007-08-30 2013-07-17 腾讯科技（深圳）有限公司 聊天方法、聊天室客户端、系统管理后台及服务器
US8638363B2 (en) 2009-02-18 2014-01-28 Google Inc. Automatically capturing information, such as capturing information using a document-aware device
US20110145068A1 (en) 2007-09-17 2011-06-16 King Martin T Associating rendered advertisements with digital content
US8082151B2 (en) 2007-09-18 2011-12-20 At&T Intellectual Property I, Lp System and method of generating responses to text-based messages
US20090119584A1 (en) 2007-11-02 2009-05-07 Steve Herbst Software Tool for Creating Outlines and Mind Maps that Generates Subtopics Automatically
KR101230705B1 (ko) 2007-12-17 2013-02-07 플레이 메가폰 유저와 대화식 시스템 간의 상호작용을 관리하는 시스템 및 방법
CN101983396B (zh) 2008-03-31 2014-07-09 皇家飞利浦电子股份有限公司 基于用户指令修改表示的方法
US20090282114A1 (en) 2008-05-08 2009-11-12 Junlan Feng System and method for generating suggested responses to an email
US20090327436A1 (en) 2008-06-30 2009-12-31 Chen Shihn-Cheng Instant messaging network control module
US8166019B1 (en) 2008-07-21 2012-04-24 Sprint Communications Company L.P. Providing suggested actions in response to textual communications
JP5273712B2 (ja) 2008-08-11 2013-08-28 シャープ株式会社 情報処理装置、情報処理方法および情報処理プログラム
US8805110B2 (en) 2008-08-19 2014-08-12 Digimarc Corporation Methods and systems for content processing
EP2321948B1 (en) 2008-08-25 2017-11-15 Orange Method to identify and transfer to a wireless device actionable items based on user selected content
US8391618B1 (en) 2008-09-19 2013-03-05 Adobe Systems Incorporated Semantic image classification and search
US8055710B2 (en) 2008-09-24 2011-11-08 International Business Machines Corporation System, method and computer program product for intelligent multi-person chat history injection
USD599363S1 (en) 2008-11-24 2009-09-01 Microsoft Corporation Transitional cursor user interface for a portion of a display screen
USD611053S1 (en) 2008-11-24 2010-03-02 Microsoft Corporation Transitional user interface for a portion of a display screen
US20130036162A1 (en) 2009-02-10 2013-02-07 Mikekoenigs.Com, Inc. Automated Communication Techniques
US20100228590A1 (en) 2009-03-03 2010-09-09 International Business Machines Corporation Context-aware electronic social networking
JP4739438B2 (ja) 2009-03-26 2011-08-03 株式会社エヌ・ティ・ティ・ドコモ 通信端末及びメール返信方法
US20100251158A1 (en) 2009-03-30 2010-09-30 Avaya Inc. System and method for graphically managing communication sessions
US9195898B2 (en) 2009-04-14 2015-11-24 Qualcomm Incorporated Systems and methods for image recognition using mobile devices
US20120131520A1 (en) 2009-05-14 2012-05-24 Tang ding-yuan Gesture-based Text Identification and Selection in Images
US9782527B2 (en) 2009-05-27 2017-10-10 Tc1 Llc Monitoring of redundant conductors
US9043407B1 (en) 2009-06-12 2015-05-26 Avaya Inc. Interactive user interface to communication-enabled business process platforms method and apparatus
USD651609S1 (en) 2009-06-26 2012-01-03 Microsoft Corporation Display screen with an animated image
CN102667754A (zh) 2009-07-02 2012-09-12 乐宅睦有限公司 用于增强数字内容的系统和方法
US8572084B2 (en) 2009-07-28 2013-10-29 Fti Consulting, Inc. System and method for displaying relationships between electronically stored information to provide classification suggestions via nearest neighbor
US9128610B2 (en) 2009-09-30 2015-09-08 At&T Mobility Ii Llc Virtual predictive keypad
US8831279B2 (en) 2011-03-04 2014-09-09 Digimarc Corporation Smartphone-based methods and systems
US8121618B2 (en) 2009-10-28 2012-02-21 Digimarc Corporation Intuitive computing methods and systems
US8400548B2 (en) 2010-01-05 2013-03-19 Apple Inc. Synchronized, interactive augmented reality displays for multifunction devices
USD624927S1 (en) 2010-01-19 2010-10-05 Microsoft Corporation User interface for a portion of a display screen
US8650210B1 (en) 2010-02-09 2014-02-11 Google Inc. Identifying non-search actions based on a search query
US8782556B2 (en) * 2010-02-12 2014-07-15 Microsoft Corporation User-centric soft keyboard predictive technologies
US8655965B2 (en) 2010-03-05 2014-02-18 Qualcomm Incorporated Automated messaging response in wireless communication systems
US8266109B1 (en) 2010-03-09 2012-09-11 Symantec Corporation Performance of scanning containers for archiving
JP2011211696A (ja) 2010-03-10 2011-10-20 Nikon Corp 画像データ処理システム、画像データ処理プログラム、及び画像データ処理装置
JP5733907B2 (ja) 2010-04-07 2015-06-10 キヤノン株式会社 画像処理装置、画像処理方法、コンピュータプログラム
US20110252207A1 (en) 2010-04-08 2011-10-13 Oracle International Corporation Dynamic content archiving
US9929982B2 (en) 2010-04-08 2018-03-27 Microsoft Technology Licensing, Llc Designating automated agents as friends in a social network service
US20170098122A1 (en) 2010-06-07 2017-04-06 Affectiva, Inc. Analysis of image content with associated manipulation of expression presentation
USD648343S1 (en) 2010-06-24 2011-11-08 Microsoft Corporation Display screen with user interface
USD648735S1 (en) 2010-06-25 2011-11-15 Microsoft Corporation Display screen with animated user interface
US20120030289A1 (en) 2010-07-30 2012-02-02 Avaya Inc. System and method for multi-model, context-sensitive, real-time collaboration
US8781152B2 (en) 2010-08-05 2014-07-15 Brian Momeyer Identifying visual media content captured by camera-enabled mobile device
US9936333B2 (en) 2010-08-10 2018-04-03 Microsoft Technology Licensing, Llc Location and contextual-based mobile application promotion and delivery
KR101722687B1 (ko) 2010-08-10 2017-04-04 삼성전자주식회사 객체간 또는 객체와 유저간에 정보 제공 방법, 그 방법을 수행할 수 있는 유저 디바이스, 및 저장 매체
US8966590B2 (en) 2010-08-17 2015-02-24 Facebook, Inc. Managing social network accessibility based on age
US9262517B2 (en) 2010-08-18 2016-02-16 At&T Intellectual Property I, L.P. Systems and methods for social media data mining
US8566911B2 (en) 2010-10-06 2013-10-22 Blackberry Limited Method of obtaining authorization for accessing a service
KR101753031B1 (ko) 2010-11-15 2017-06-30 엘지전자 주식회사 이동 단말기 및 이것의 메타데이터 설정 방법
KR101060753B1 (ko) 2011-01-04 2011-08-31 (주)올라웍스 입력 이미지에 포함된 객체에 대한 콜렉션을 수행할 수 있도록 지원하기 위한 방법, 단말 장치 및 컴퓨터 판독 가능한 기록 매체
US20120179717A1 (en) * 2011-01-11 2012-07-12 Sony Corporation System and method for effectively providing entertainment recommendations to device users
US8688698B1 (en) 2011-02-11 2014-04-01 Google Inc. Automatic text suggestion
EP2678822A4 (en) 2011-02-23 2014-09-10 Bottlenose Inc System and method for message analysis in a network or between networks
US20130262574A1 (en) 2011-03-15 2013-10-03 Gabriel Cohen Inline User Addressing in Chat Sessions
US8938669B1 (en) 2011-03-15 2015-01-20 Google Inc. Inline user addressing in chat and document editing sessions
US8849931B2 (en) 2011-03-15 2014-09-30 Idt Messaging, Llc Linking context-based information to text messages
US8554701B1 (en) 2011-03-18 2013-10-08 Amazon Technologies, Inc. Determining sentiment of sentences from customer reviews
JP2012221480A (ja) 2011-04-06 2012-11-12 L Is B Corp メッセージ処理システム
WO2012150602A1 (en) 2011-05-03 2012-11-08 Yogesh Chunilal Rathod A system and method for dynamically monitoring, recording, processing, attaching dynamic, contextual & accessible active links & presenting of physical or digital activities, actions, locations, logs, life stream, behavior & status
EP2523436A1 (en) 2011-05-11 2012-11-14 Alcatel Lucent Mobile device and method of managing applications for a mobile device
USD658677S1 (en) 2011-05-27 2012-05-01 Microsoft Corporation Display screen with animated user interface
USD658678S1 (en) 2011-05-27 2012-05-01 Microsoft Corporation Display screen with animated user interface
USD658201S1 (en) 2011-05-27 2012-04-24 Microsoft Corporation Display screen with animated user interface
US8832284B1 (en) 2011-06-16 2014-09-09 Google Inc. Virtual socializing
US8589407B2 (en) 2011-06-17 2013-11-19 Google Inc. Automated generation of suggestions for personalized reactions in a social network
US8881250B2 (en) 2011-06-17 2014-11-04 Ebay Inc. Passporting credentials between a mobile app and a web browser
US8700480B1 (en) 2011-06-20 2014-04-15 Amazon Technologies, Inc. Extracting quotes from customer reviews regarding collections of items
US9588668B2 (en) 2011-07-21 2017-03-07 Imerj, Llc Methods of displaying a second view
US9245253B2 (en) 2011-08-19 2016-01-26 Disney Enterprises, Inc. Soft-sending chat messages
US8659667B2 (en) 2011-08-29 2014-02-25 Panasonic Corporation Recipe based real-time assistance for digital image capture and other consumer electronics devices
US9179278B2 (en) 2011-09-01 2015-11-03 Qualcomm Incorporated Systems and methods involving augmented menu using mobile device
US10102546B2 (en) 2011-09-15 2018-10-16 Stephan HEATH System and method for tracking, utilizing predicting, and implementing online consumer browsing behavior, buying patterns, social networking communications, advertisements and communications, for online coupons, products, goods and services, auctions, and service providers using geospatial mapping technology, and social networking
EP3200185A1 (en) 2011-09-30 2017-08-02 Apple Inc. Using context information to facilitate processing of commands in a virtual assistant
US20180032997A1 (en) 2012-10-09 2018-02-01 George A. Gordon System, method, and computer program product for determining whether to prompt an action by a platform in connection with a mobile device
KR101521332B1 (ko) 2011-11-08 2015-05-20 주식회사 다음카카오 인스턴트 메시징 서비스 및 인스턴트 메시징 서비스로부터 확장된 복수의 서비스들을 제공하는 방법
US9697016B2 (en) 2011-11-15 2017-07-04 Microsoft Technology Licensing, Llc Search augmented menu and configuration for computer applications
USD673172S1 (en) 2011-11-21 2012-12-25 Microsoft Corporation Display screen with animated graphical user interface
KR101402506B1 (ko) 2011-12-01 2014-06-03 라인 가부시키가이샤 인스턴트 메시징 어플리케이션을 이용한 대화형 정보제공 시스템 및 방법
USD701228S1 (en) 2012-01-06 2014-03-18 Samsung Electronics Co., Ltd. Display screen or portion thereof with transitional graphical user interface
USD699744S1 (en) 2012-01-06 2014-02-18 Microsoft Corporation Display screen with an animated graphical user interface
AU344579S (en) 2012-01-09 2012-09-27 Samsung Electronics Co Ltd Display screen for an electronic device
USD705802S1 (en) 2012-02-07 2014-05-27 Microsoft Corporation Display screen with animated graphical user interface
USD705251S1 (en) 2012-02-09 2014-05-20 Microsoft Corporation Display screen with animated graphical user interface
US9306878B2 (en) 2012-02-14 2016-04-05 Salesforce.Com, Inc. Intelligent automated messaging for computer-implemented devices
GB2499395A (en) 2012-02-14 2013-08-21 British Sky Broadcasting Ltd Search method
US20130218885A1 (en) * 2012-02-22 2013-08-22 Salesforce.Com, Inc. Systems and methods for context-aware message tagging
USD699739S1 (en) 2012-02-22 2014-02-18 Microsoft Corporation Display screen with animated graphical user interface
USD701527S1 (en) 2012-02-23 2014-03-25 Htc Corporation Display screen with transitional graphical user interface
USD701528S1 (en) 2012-02-24 2014-03-25 Htc Corporation Display screen with transitional graphical user interface
US8620021B2 (en) 2012-03-29 2013-12-31 Digimarc Corporation Image-related methods and arrangements
US8855430B1 (en) 2012-05-30 2014-10-07 Google Inc. Refining image annotations
USD705244S1 (en) 2012-06-20 2014-05-20 Microsoft Corporation Display screen with animated graphical user interface
US20130346235A1 (en) 2012-06-20 2013-12-26 Ebay, Inc. Systems, Methods, and Computer Program Products for Caching of Shopping Items
US9191786B2 (en) 2012-06-27 2015-11-17 At&T Intellectual Property I, L.P. Method and apparatus for generating a suggested message to be sent over a network
US9412136B2 (en) 2012-07-09 2016-08-09 Facebook, Inc. Creation of real-time conversations based on social location information
KR20140011073A (ko) 2012-07-17 2014-01-28 삼성전자주식회사 텍스트 추천 방법 및 장치
US9019415B2 (en) 2012-07-26 2015-04-28 Qualcomm Incorporated Method and apparatus for dual camera shutter
US9195645B2 (en) 2012-07-30 2015-11-24 Microsoft Technology Licensing, Llc Generating string predictions using contexts
KR101899817B1 (ko) 2012-08-01 2018-09-19 엘지전자 주식회사 이동단말기 및 그 제어 방법
USD695755S1 (en) 2012-08-06 2013-12-17 Samsung Electronics Co., Ltd. TV monitor with graphical user interface
US20140047413A1 (en) 2012-08-09 2014-02-13 Modit, Inc. Developing, Modifying, and Using Applications
US20140052540A1 (en) * 2012-08-20 2014-02-20 Giridhar Rajaram Providing content using inferred topics extracted from communications in a social networking system
USD706802S1 (en) 2012-08-28 2014-06-10 Samsung Electronics Co., Ltd. Portable electronic device displaying transitional graphical user interface
KR102068604B1 (ko) 2012-08-28 2020-01-22 삼성전자 주식회사 휴대단말기의 문자 인식장치 및 방법
US9218333B2 (en) 2012-08-31 2015-12-22 Microsoft Technology Licensing, Llc Context sensitive auto-correction
JP6160996B2 (ja) 2012-09-12 2017-07-12 パナソニックＩｐマネジメント株式会社 撮像装置
US20140088954A1 (en) 2012-09-27 2014-03-27 Research In Motion Limited Apparatus and method pertaining to automatically-suggested emoticons
US10691743B2 (en) 2014-08-05 2020-06-23 Sri International Multi-dimensional realization of visual content of an image collection
US9299060B2 (en) 2012-10-12 2016-03-29 Google Inc. Automatically suggesting groups based on past user interaction
KR20140052155A (ko) 2012-10-19 2014-05-07 삼성전자주식회사 디스플레이 장치, 디스플레이 장치 제어 방법 및 디스플레이 장치의 제어를 위한 정보처리장치
USD714821S1 (en) 2012-10-24 2014-10-07 Microsoft Corporation Display screen with animated graphical user interface
US20150286371A1 (en) 2012-10-31 2015-10-08 Aniways Advertising Solutions Ltd. Custom emoticon generation
US20140156801A1 (en) * 2012-12-04 2014-06-05 Mobitv, Inc. Cowatching and connected platforms using a push architecture
US9244905B2 (en) 2012-12-06 2016-01-26 Microsoft Technology Licensing, Llc Communication context based predictive-text suggestion
US20140164506A1 (en) 2012-12-10 2014-06-12 Rawllin International Inc. Multimedia message having portions of networked media content
US20140171133A1 (en) 2012-12-18 2014-06-19 Google Inc. Query response
CN103067490B (zh) 2012-12-26 2015-11-25 腾讯科技（深圳）有限公司 移动终端通讯会话的通知方法、终端、服务器及系统
GB201322037D0 (en) 2013-12-12 2014-01-29 Touchtype Ltd System and method for inputting images/labels into electronic devices
US9020956B1 (en) 2012-12-31 2015-04-28 Google Inc. Sentiment and topic based content determination methods and systems
US20140189538A1 (en) 2012-12-31 2014-07-03 Motorola Mobility Llc Recommendations for Applications Based on Device Context
US8930481B2 (en) 2012-12-31 2015-01-06 Huawei Technologies Co., Ltd. Message processing method, terminal and system
US9374327B2 (en) 2013-01-08 2016-06-21 Vmware, Inc. Intelligent chat system
KR20140091633A (ko) 2013-01-11 2014-07-22 삼성전자주식회사 모바일 장치에서의 상황 인지에 따른 추천 항목을 제공하기 위한 방법 및 이를 위한 모바일 장치
KR101821358B1 (ko) 2013-01-22 2018-01-25 네이버 주식회사 다자간 메신저 서비스를 제공하는 방법 및 시스템
JP6255646B2 (ja) 2013-03-04 2018-01-10 株式会社Ｌ ｉｓ Ｂ メッセージシステム
USD704726S1 (en) 2013-03-04 2014-05-13 Roger Leslie Maxwell Display screen or portion thereof with animated graphical user interface
US20140344058A1 (en) 2013-03-15 2014-11-20 Fision Holdings, Inc Systems and methods for distributed marketing automation
US20140298364A1 (en) 2013-03-26 2014-10-02 Rawllin International Inc. Recommendations for media content based on emotion
US8825474B1 (en) 2013-04-16 2014-09-02 Google Inc. Text suggestion output using past interaction data
US9177318B2 (en) 2013-04-22 2015-11-03 Palo Alto Research Center Incorporated Method and apparatus for customizing conversation agents based on user characteristics using a relevance score for automatic statements, and a response prediction function
US9923849B2 (en) 2013-05-09 2018-03-20 Ebay Inc. System and method for suggesting a phrase based on a context
WO2014200621A1 (en) 2013-06-13 2014-12-18 Evernote Corporation Initializing chat sessions by pointing to content
US10599765B2 (en) 2013-06-27 2020-03-24 Avaya Inc. Semantic translation model training
ES2707967T3 (es) 2013-07-16 2019-04-08 Pinterest Inc Controles de menú contextual basado en objetos
US9330110B2 (en) 2013-07-17 2016-05-03 Xerox Corporation Image search system and method for personalized photo applications using semantic networks
US9794198B2 (en) 2013-07-19 2017-10-17 Tencent Technology (Shenzhen) Company Limited Methods and systems for creating auto-reply messages
US10162884B2 (en) 2013-07-23 2018-12-25 Conduent Business Services, Llc System and method for auto-suggesting responses based on social conversational contents in customer care services
US9161188B2 (en) 2013-08-22 2015-10-13 Yahoo! Inc. System and method for automatically suggesting diverse and personalized message completions
CN104035947B (zh) * 2013-09-16 2016-04-13 腾讯科技（深圳）有限公司 兴趣点推荐方法及装置、获取推荐兴趣点的方法及装置
EP2852105A1 (en) 2013-09-20 2015-03-25 Ignazio Di Tocco Computer system and related process supporting the communication of users located in the same geographical area, in order to establish a starting contact leading to a personal communication
US9401881B2 (en) 2013-09-26 2016-07-26 International Business Machines Corporation Automatic question generation and answering based on monitored messaging sessions
US9329692B2 (en) 2013-09-27 2016-05-03 Microsoft Technology Licensing, Llc Actionable content displayed on a touch screen
US20150100537A1 (en) 2013-10-03 2015-04-09 Microsoft Corporation Emoji for Text Predictions
US8996639B1 (en) 2013-10-15 2015-03-31 Google Inc. Predictive responses to incoming communications
US20150127753A1 (en) 2013-11-04 2015-05-07 Meemo, Llc Word Recognition and Ideograph or In-App Advertising System
WO2015089483A1 (en) 2013-12-12 2015-06-18 Mobile Iron, Inc. Application synchornization
KR20150071768A (ko) 2013-12-18 2015-06-29 에스케이하이닉스 주식회사 이미지 센서 및 그 제조방법
US9661067B2 (en) 2013-12-23 2017-05-23 24/7 Customer, Inc. Systems and methods for facilitating dialogue mining
US9519408B2 (en) 2013-12-31 2016-12-13 Google Inc. Systems and methods for guided user actions
US9817813B2 (en) 2014-01-08 2017-11-14 Genesys Telecommunications Laboratories, Inc. Generalized phrases in automatic speech recognition systems
US20150207765A1 (en) 2014-01-17 2015-07-23 Nathaniel Brantingham Messaging Service with Conversation Suggestions
US9721183B2 (en) 2014-01-31 2017-08-01 Hulu, LLC Intelligent determination of aesthetic preferences based on user history and properties
US9515968B2 (en) 2014-02-05 2016-12-06 Facebook, Inc. Controlling access to ideograms
WO2015120019A1 (en) 2014-02-10 2015-08-13 Google Inc. Smart camera user interface
CN104836720B (zh) * 2014-02-12 2022-02-25 北京三星通信技术研究有限公司 交互式通信中进行信息推荐的方法及装置
US10095748B2 (en) 2014-03-03 2018-10-09 Microsoft Technology Licensing, Llc Personalized information query suggestions
CN103841007B (zh) 2014-03-04 2017-05-10 腾讯科技（深圳）有限公司 网络游戏系统中的数据处理方法、装置及系统
KR102106787B1 (ko) 2014-03-17 2020-05-06 에스케이텔레콤 주식회사 메신저 프로그램과 어플리케이션 연동 방법, 이를 위한 장치 및 시스템
CN104951428B (zh) 2014-03-26 2019-04-16 阿里巴巴集团控股有限公司 用户意图识别方法及装置
US9544257B2 (en) 2014-04-04 2017-01-10 Blackberry Limited System and method for conducting private messaging
US10482163B2 (en) 2014-04-23 2019-11-19 Klickafy, Llc Clickable emoji
CN103995872B (zh) 2014-05-21 2017-04-05 王青 一种在应用中基于场景展开讨论与聊天的方法与系统
US10255449B2 (en) 2014-05-30 2019-04-09 Apple Inc. Permission request
US10445396B2 (en) 2014-05-31 2019-10-15 Apple Inc. Device, method, and graphical user interface for extending functionality of a host application to another application
US9380010B2 (en) 2014-06-03 2016-06-28 International Business Machines Corporation Conversation branching for more efficient resolution
US9858295B2 (en) 2014-06-24 2018-01-02 Google Llc Ranking and selecting images for display from a set of images
EP3165012A1 (en) 2014-07-03 2017-05-10 Nuance Communications, Inc. System and method for suggesting actions based upon incoming messages
US9043196B1 (en) 2014-07-07 2015-05-26 Machine Zone, Inc. Systems and methods for identifying and suggesting emoticons
US9082018B1 (en) 2014-09-30 2015-07-14 Google Inc. Method and system for retroactively changing a display characteristic of event indicators on an event timeline
US9990105B2 (en) 2014-07-08 2018-06-05 Verizon Patent And Licensing Inc. Accessible contextual controls within a graphical user interface
US20160043817A1 (en) 2014-07-18 2016-02-11 RSS Technologies, LLC Methods and apparatus for locality based broadcasting
EP3757928A1 (en) 2014-07-31 2020-12-30 Samsung Electronics Co., Ltd. Message service providing device and method of providing content via the same
CN104202718A (zh) 2014-08-05 2014-12-10 百度在线网络技术（北京）有限公司 一种向用户提供信息的方法与装置
US10218652B2 (en) 2014-08-08 2019-02-26 Mastercard International Incorporated Systems and methods for integrating a chat function into an e-reader application
US9965559B2 (en) 2014-08-21 2018-05-08 Google Llc Providing automatic actions for mobile onscreen content
US9705832B2 (en) 2014-08-27 2017-07-11 Lenovo (Singapore) Pte. Ltd. Context-aware aggregation of text-based messages
US10447621B2 (en) 2014-09-04 2019-10-15 Microsoft Technology Licensing, Llc App powered extensibility of messages on an existing messaging service
US10146748B1 (en) 2014-09-10 2018-12-04 Google Llc Embedding location information in a media collaboration using natural language processing
WO2016072117A1 (ja) 2014-11-07 2016-05-12 ソニー株式会社 情報処理装置、制御方法、および記憶媒体
US20160140477A1 (en) 2014-11-13 2016-05-19 Xerox Corporation Methods and systems for assigning tasks to workers
US9569728B2 (en) 2014-11-14 2017-02-14 Bublup Technologies, Inc. Deriving semantic relationships based on empirical organization of content by users
CN105786455B (zh) 2014-12-17 2020-02-18 深圳市腾讯计算机系统有限公司 一种数据处理方法、装置及终端
US20160179816A1 (en) 2014-12-22 2016-06-23 Quixey, Inc. Near Real Time Auto-Suggest Search Results
US9727218B2 (en) 2015-01-02 2017-08-08 Microsoft Technology Licensing, Llc Contextual browser frame and entry box placement
KR101634086B1 (ko) 2015-01-19 2016-07-08 주식회사 엔씨소프트 감정 분석을 통한 스티커 추천 방법 및 시스템
KR20160089152A (ko) 2015-01-19 2016-07-27 주식회사 엔씨소프트 화행 분석을 통한 스티커 추천 방법 및 시스템
US20160224524A1 (en) 2015-02-03 2016-08-04 Nuance Communications, Inc. User generated short phrases for auto-filling, automatically collected during normal text use
US20160226804A1 (en) 2015-02-03 2016-08-04 Google Inc. Methods, systems, and media for suggesting a link to media content
US9661386B2 (en) 2015-02-11 2017-05-23 Google Inc. Methods, systems, and media for presenting a suggestion to watch videos
US10079785B2 (en) 2015-02-12 2018-09-18 Google Llc Determining reply content for a reply to an electronic communication
US20160284011A1 (en) 2015-03-25 2016-09-29 Facebook, Inc. Techniques for social messaging authorization and customization
US10353542B2 (en) 2015-04-02 2019-07-16 Facebook, Inc. Techniques for context sensitive illustrated graphical user interface elements
US10965622B2 (en) 2015-04-16 2021-03-30 Samsung Electronics Co., Ltd. Method and apparatus for recommending reply message
US9703541B2 (en) 2015-04-28 2017-07-11 Google Inc. Entity action suggestion on a mobile device
US9883358B2 (en) 2015-05-08 2018-01-30 Blackberry Limited Electronic device and method of determining suggested responses to text-based communications
US10909329B2 (en) 2015-05-21 2021-02-02 Baidu Usa Llc Multilingual image question answering
US10504509B2 (en) 2015-05-27 2019-12-10 Google Llc Providing suggested voice-based action queries
US10091140B2 (en) 2015-05-31 2018-10-02 Microsoft Technology Licensing, Llc Context-sensitive generation of conversational responses
KR20160148260A (ko) 2015-06-16 2016-12-26 삼성전자주식회사 전자 장치 및 이의 제어 방법
US10274911B2 (en) 2015-06-25 2019-04-30 Intel Corporation Conversational interface for matching text of spoken input based on context model
US10042866B2 (en) 2015-06-30 2018-08-07 Adobe Systems Incorporated Searching untagged images with text-based queries
US9712466B2 (en) 2015-11-10 2017-07-18 Wrinkl, Inc. Integrating actionable objects into an on-line chat communications platform
EP4645114A2 (en) 2015-07-15 2025-11-05 rewardStyle, Inc. Systems and methods for screenshot linking
US20170031575A1 (en) 2015-07-28 2017-02-02 Microsoft Technology Licensing, Llc Tailored computing experience based on contextual signals
CN105183276A (zh) 2015-08-19 2015-12-23 小米科技有限责任公司 在聊天界面实现游戏的方法、装置及终端设备
US10445425B2 (en) 2015-09-15 2019-10-15 Apple Inc. Emoji and canned responses
US9467435B1 (en) 2015-09-15 2016-10-11 Mimecast North America, Inc. Electronic message threat protection system for authorized users
US11025569B2 (en) 2015-09-30 2021-06-01 Apple Inc. Shared content presentation with integrated messaging
US10789525B2 (en) 2015-10-02 2020-09-29 Adobe Inc. Modifying at least one attribute of an image with at least one attribute extracted from another image
KR20170048964A (ko) 2015-10-27 2017-05-10 라인 가부시키가이샤 메시지 제공 방법, 장치, 및 컴퓨터 프로그램과, 표시 제어 방법, 장치 및 컴퓨터 프로그램
CN105262675A (zh) 2015-10-29 2016-01-20 北京奇虎科技有限公司 基于电子书进行的聊天控制方法和装置
KR102393928B1 (ko) 2015-11-10 2022-05-04 삼성전자주식회사 응답 메시지를 추천하는 사용자 단말 장치 및 그 방법
US9633048B1 (en) 2015-11-16 2017-04-25 Adobe Systems Incorporated Converting a text sentence to a series of images
US10129193B2 (en) 2015-11-17 2018-11-13 International Business Machines Corporation Identifying relevant content contained in message streams that appear to be irrelevant
US20170147202A1 (en) 2015-11-24 2017-05-25 Facebook, Inc. Augmenting text messages with emotion information
KR102427833B1 (ko) 2015-11-30 2022-08-02 삼성전자주식회사 사용자 단말장치 및 디스플레이 방법
US20170171117A1 (en) 2015-12-10 2017-06-15 International Business Machines Corporation Message Suggestion Using Dynamic Information
WO2017112786A1 (en) 2015-12-21 2017-06-29 Google Inc. Automatic suggestions for message exchange threads
CN108476164B (zh) 2015-12-21 2021-10-08 谷歌有限责任公司 在消息传送应用中自动地提供机器人服务的方法
US10732783B2 (en) 2015-12-28 2020-08-04 Microsoft Technology Licensing, Llc Identifying image comments from similar images
KR101712180B1 (ko) 2015-12-29 2017-03-06 라인 가부시키가이샤 메시지를 송수신하기 위한 컴퓨터로 읽을 수 있는 기록매체에 저장된 프로그램, 방법 및 장치
US9560152B1 (en) 2016-01-27 2017-01-31 International Business Machines Corporation Personalized summary of online communications
US20170250935A1 (en) 2016-02-25 2017-08-31 Facebook, Inc. Techniques for messaging bot app interactions
US11477139B2 (en) 2016-02-25 2022-10-18 Meta Platforms, Inc. Techniques for messaging bot rich communication
US20170250930A1 (en) 2016-02-29 2017-08-31 Outbrain Inc. Interactive content recommendation personalization assistant
US20170288942A1 (en) 2016-03-30 2017-10-05 Microsoft Technology Licensing, Llc Portal for Provisioning Autonomous Software Agents
US10831802B2 (en) 2016-04-11 2020-11-10 Facebook, Inc. Techniques to respond to user requests using natural-language machine learning based on example conversations
US10452671B2 (en) 2016-04-26 2019-10-22 Facebook, Inc. Recommendations from comments on online social networks
US9866693B2 (en) 2016-05-06 2018-01-09 Genesys Telecommunications Laboratories, Inc. System and method for monitoring progress of automated chat conversations
US20170344224A1 (en) 2016-05-27 2017-11-30 Nuance Communications, Inc. Suggesting emojis to users for insertion into text-based messages
CN105898627B (zh) 2016-05-31 2019-04-12 北京奇艺世纪科技有限公司 一种视频播放方法及装置
US10505872B2 (en) 2016-06-12 2019-12-10 Apple Inc. Messaging application interacting with one or more extension applications
US10852912B2 (en) 2016-06-12 2020-12-01 Apple Inc. Image creation app in messaging app
US11088973B2 (en) 2016-06-12 2021-08-10 Apple Inc. Conversion of text relating to media content and media extension apps
US10785175B2 (en) 2016-06-12 2020-09-22 Apple Inc. Polling extension application for interacting with a messaging application
US10368208B2 (en) 2016-06-12 2019-07-30 Apple Inc. Layers in messaging applications
US20170359283A1 (en) 2016-06-12 2017-12-14 Apple Inc. Music creation app in messaging app
US9990128B2 (en) 2016-06-12 2018-06-05 Apple Inc. Messaging application interacting with one or more extension applications
US10554599B2 (en) 2016-06-12 2020-02-04 Apple Inc. Conversion of detected URL in text message
US10194288B2 (en) 2016-06-12 2019-01-29 Apple Inc. Sticker distribution system for messaging apps
US10595169B2 (en) 2016-06-12 2020-03-17 Apple Inc. Message extension app store
US20170366479A1 (en) 2016-06-20 2017-12-21 Microsoft Technology Licensing, Llc Communication System
US10254935B2 (en) 2016-06-29 2019-04-09 Google Llc Systems and methods of providing content selection
US10515393B2 (en) 2016-06-30 2019-12-24 Paypal, Inc. Image data detection for micro-expression analysis and targeted data services
WO2018009231A1 (en) 2016-07-08 2018-01-11 Asapp, Inc. Automatically responding to a request of a user
US20180032499A1 (en) 2016-07-28 2018-02-01 Google Inc. Automatically Generating Spelling Suggestions and Corrections Based on User Context
US10049310B2 (en) 2016-08-30 2018-08-14 International Business Machines Corporation Image text analysis for identifying hidden text
KR20180026983A (ko) 2016-09-05 2018-03-14 삼성전자주식회사 전자 장치 및 그 제어 방법
KR101961754B1 (ko) 2016-09-20 2019-03-25 구글 엘엘씨 봇 상호작용
US10015124B2 (en) 2016-09-20 2018-07-03 Google Llc Automatic response suggestions based on images received in messaging applications
DE112017003594T5 (de) 2016-09-20 2019-04-25 Google Llc Bot, der Genehmigung zum Zugriff auf Daten anfordert
CN109952572B (zh) 2016-09-20 2023-11-24 谷歌有限责任公司 基于消息贴纸的建议响应
US11176931B2 (en) 2016-09-23 2021-11-16 Microsoft Technology Licensing, Llc Conversational bookmarks
CN106484831A (zh) 2016-09-29 2017-03-08 百度在线网络技术（北京）有限公司 搜索系统、方法和装置
US10416846B2 (en) 2016-11-12 2019-09-17 Google Llc Determining graphical element(s) for inclusion in an electronic communication
US20180196854A1 (en) 2017-01-11 2018-07-12 Google Inc. Application extension for generating automatic search queries
US10146768B2 (en) 2017-01-25 2018-12-04 Google Llc Automatic suggested responses to images received in messages using language model
US10229427B2 (en) 2017-04-10 2019-03-12 Wildfire Systems, Inc. Virtual keyboard trackable referral system
US20180316637A1 (en) 2017-05-01 2018-11-01 Microsoft Technology Licensing, Llc Conversation lens for context
IL270266B2 (en) 2017-05-05 2024-04-01 Liveperson Inc Dynamic response prediction for improved bot task processing
US10891485B2 (en) 2017-05-16 2021-01-12 Google Llc Image archival based on image categories
US10827319B2 (en) 2017-06-02 2020-11-03 Apple Inc. Messaging system interacting with dynamic extension app
US10404636B2 (en) 2017-06-15 2019-09-03 Google Llc Embedded programs and interfaces for chat conversations
US10348658B2 (en) 2017-06-15 2019-07-09 Google Llc Suggested items for use with embedded applications in chat conversations

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140237057A1 (en) * 2013-02-21 2014-08-21 Genesys Telecommunications Laboratories, Inc. System and method for processing private messages in a contact center
US20150178388A1 (en) * 2013-12-19 2015-06-25 Adobe Systems Incorporated Interactive communication augmented with contextual information
US20150302301A1 (en) * 2014-04-22 2015-10-22 Google Inc. Automatic actions based on contextual replies

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE102016125509A1 (de) * 2016-05-17 2017-11-23 Google Inc. Einbindung von auswählbaren Anwendungsverknüpfungen in Nachrichtenaustausch-Threads
US10263933B2 (en) 2016-05-17 2019-04-16 Google Llc Incorporating selectable application links into message exchange threads
US10291565B2 (en) 2016-05-17 2019-05-14 Google Llc Incorporating selectable application links into conversations with personal assistant modules
US10601755B2 (en) 2016-05-17 2020-03-24 Google Llc Incorporating selectable application links into conversations with personal assistant modules
US11057333B2 (en) 2016-05-17 2021-07-06 Google Llc Incorporating selectable application links into message exchange threads
US11176520B2 (en) 2019-04-18 2021-11-16 Microsoft Technology Licensing, Llc Email content modification system

Also Published As

Publication number Publication date
JP6638087B2 (ja) 2020-01-29
US11418471B2 (en) 2022-08-16
CN108781175A (zh) 2018-11-09
US20200092243A1 (en) 2020-03-19
US20170180294A1 (en) 2017-06-22
US10530723B2 (en) 2020-01-07
JP2019508820A (ja) 2019-03-28
CN108781175B (zh) 2021-09-21
KR20180093040A (ko) 2018-08-20
EP3395018A1 (en) 2018-10-31

Similar Documents

Publication Publication Date Title
US11418471B2 (en) Automatic suggestions for message exchange threads
US11502975B2 (en) Automatic suggestions and other content for messaging applications
US11303590B2 (en) Suggested responses based on message stickers
US9946782B2 (en) System and method for message clustering
US8386506B2 (en) System and method for context enhanced messaging
US9747583B2 (en) Presenting entity profile information to a user of a computing device
US10540666B2 (en) Method and system for updating an intent space and estimating intent based on an intent space
US11232522B2 (en) Methods, systems and techniques for blending online content from multiple disparate content sources including a personal content source or a semi-personal content source
US10417206B2 (en) Method and system for associating data from different sources to generate a person-centric space
US10387432B2 (en) Methods, systems and techniques for ranking blended content retrieved from multiple disparate content sources
US20170097959A1 (en) Method and system for searching in a person-centric space

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 16825663

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2018551908

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 20187019756

Country of ref document: KR

Kind code of ref document: A

WWE Wipo information: entry into national phase

Ref document number: 1020187019756

Country of ref document: KR

WWE Wipo information: entry into national phase

Ref document number: 2016825663

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 2016825663

Country of ref document: EP

Effective date: 20180723