CN111914070A

CN111914070A - Intelligent information prompting assistant system, information prompting method and terminal equipment

Info

Publication number: CN111914070A
Application number: CN202010389181.7A
Authority: CN
Inventors: 丁翔; 王瑞鑫; 吴琨
Original assignee: Shanghai Chule Cootek Information Technology Co Ltd
Current assignee: Shanghai Chule Cootek Information Technology Co Ltd
Priority date: 2019-05-09
Filing date: 2020-05-09
Publication date: 2020-11-10

Abstract

The intelligent information prompting assistant system, the information prompting method of the intelligent robot assistant system and the terminal device provide services from a third view angle, wherein the intelligent information prompting assistant system comprises: the monitoring device is suitable for monitoring natural interactive language of interaction between a user and an object except the intelligent information prompt assistant; the scene discrimination device is suitable for preprocessing the natural interactive language, determining the language type of the natural interactive language, judging whether information prompt service needs to be provided for the natural interactive language or not, and sending the preprocessed natural interactive language to the service device when the service needs to be provided; the service device comprises at least one service module, and the service module is suitable for selecting a corresponding knowledge base according to the language type to retrieve the preprocessed natural interactive language; the display device is suitable for displaying the retrieval result returned by the service module; and the knowledge base is suitable for storing data matched with the service provided by the service module.

Description

Intelligent information prompting assistant system, information prompting method and terminal equipment

Technical Field

The embodiment of the invention relates to an information processing technology, in particular to an intelligent information prompting assistant system, an information prompting method and terminal equipment.

Background

In recent years, the heat tide of artificial intelligence has been raised worldwide, and the application field of the intelligent assistant as artificial intelligence is rapidly developing. Most of the intelligent assistants on the market today are in the form of chat robots for human-computer interaction, such as the intelligent voice Assistant Siri of apple, Alexa of amazon, and Google Assistant of Google, etc. The user can initiate questions to the intelligent assistants through the voice input device, the intelligent assistants feed corresponding results back to the user after retrieving, for example, the user inputs 'how the weather is today' through voice, and the intelligent assistants feed corresponding weather information back to the user after inquiring; for another example, the user actively asks the intelligent assistant about "23 × 7, and the intelligent assistant feeds back the corresponding operation result to the user after operation.

It can be seen that the current intelligent assistants are used as direct communication objects of users to provide services for the users, and when the communication objects of the users are not the intelligent assistants, the intelligent assistants do not provide corresponding services any more; in addition, the current intelligent assistants focus on feeding back the corresponding results according to the contents that the user has input, and do not focus on the contents that the user wants to input. For example, when a user chats with other users about a dining topic, the user wants to search restaurant information and recommend the restaurant information to the other users, and since the chat object of the user is not an intelligent assistant and the user does not ask a question to the intelligent assistant, the intelligent assistant does not feed the restaurant information back to the user, and the user can only open other application programs, such as beauty groups, popular comments and the like, to actively search and then return a chat page to input a search result, so that the operation of the user is increased, the retrieval efficiency is reduced, and meanwhile, the input speed of the user is also reduced.

Disclosure of Invention

The embodiment of the invention provides an intelligent information prompting assistant system, an information prompting method and terminal equipment, which are used for providing corresponding services for a user from a third view angle, so that when the user interacts with an object except an intelligent information prompting assistant, corresponding prompting information is provided for the user to help the user to input quickly.

According to one aspect of the invention, an intelligent information prompting assistant system is provided, which comprises: the monitoring device is suitable for monitoring natural interactive language of interaction between a user and an object except the intelligent information prompt assistant; the scene discrimination device is suitable for preprocessing the natural interactive language, determining the language type of the natural interactive language, judging whether information prompt service needs to be provided for the natural interactive language or not, and sending the preprocessed natural interactive language to the service device when the service needs to be provided; the service device comprises at least one service module, and the service module is suitable for selecting a corresponding knowledge base according to the language type to retrieve the preprocessed natural interactive language; the display device is suitable for displaying the retrieval result returned by the service module; and the knowledge base is suitable for storing data matched with the service provided by the service module.

According to another aspect of the present invention, there is also provided an information prompting method for an intelligent robot assistant system, including: monitoring natural interactive language of interaction between a user and objects except the intelligent information prompting assistant; preprocessing the natural interactive language, and determining a language type to which the natural interactive language belongs; judging whether information prompt service needs to be provided for the natural interactive language, and sending the preprocessed natural interactive language to a service device when the service needs to be provided, wherein the service device comprises at least one service module; the service module selects a corresponding knowledge base according to the language type to retrieve the preprocessed natural interactive language; and displaying the retrieval result returned by the service module.

According to another aspect of the present invention, there is also provided a terminal device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the program executed by the processor includes the method disclosed in the embodiments of the present invention.

Through various embodiments of the invention, corresponding services are provided for the user from the third perspective, so that when the user interacts with objects except the intelligent information prompting assistant, corresponding prompting information is provided for the user to help the user to input quickly.

Drawings

FIG. 1 is a block diagram of an intelligent information prompt assistant system provided in one embodiment of the present invention;

FIG. 2 is a block diagram of an intelligent information prompt assistant system according to an embodiment of the present invention;

FIG. 3a is a diagram of an exemplary on-screen display interface provided by one embodiment of the present invention;

FIG. 3b is a diagram of an exemplary on-screen display interface provided by an embodiment of the present invention;

FIG. 3c is a diagram of an exemplary on-screen display interface provided by one embodiment of the present invention;

fig. 4 is a schematic structural diagram of a scene discrimination device according to an embodiment of the present invention;

FIG. 5a is a diagram of an exemplary on-screen display interface provided by one embodiment of the present invention;

FIG. 5b is a diagram of an exemplary on-screen display interface provided by an embodiment of the present invention;

FIG. 5c is a diagram of an exemplary on-screen display interface provided by one embodiment of the present invention;

FIG. 5d is a diagram of an exemplary on-screen display interface provided by an embodiment of the present invention;

fig. 6 is a schematic structural diagram of a service module according to an embodiment of the present invention;

fig. 7 is a schematic flow chart of an information prompting method of an intelligent robot assistant system according to an embodiment of the present invention;

FIG. 8 is a flowchart illustrating a natural interaction language processing performed by a service module according to an embodiment of the present invention;

FIG. 9 is a diagram of an exemplary on-screen display interface provided by one embodiment of the present invention;

fig. 10 is a flowchart illustrating an information prompting method of an intelligent robot assistant system according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the accompanying drawings.

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present invention. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

In this document, relational terms such as left and right, top and bottom, front and back, first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. The terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.

FIG. 1 illustrates a block diagram of an exemplary intelligent information prompt assistant system. According to some embodiments, the intelligent information prompting assistant system may be a mobile terminal including a display screen, such as a mobile phone, a smart phone, a PDA or a tablet computer, or may be other electronic devices that can interact with the internet, such as a camera, a wearable electronic device, a car navigation device, and an electronic interaction terminal disposed in a public place such as a station or a school.

The intelligent information prompt assistant system can access the network through broadband, such as ADSL, VDSL, optical fiber, wireless, cable television, satellite and the like, can also access the Internet through narrowband, such as telephone dial-up access, GPRS, 2G, 3G and the like, or can also access the telecommunication network through CDMA, 2G, 3G, 4G and the like.

According to some embodiments, the intelligent information prompt assistant system may be configured to process a natural interaction language in which a user interacts with other objects, determine whether information prompt services need to be provided, and distribute the natural interaction language to each service module in the service device for retrieval when services need to be provided, where each service module determines whether retrieval is required through confidence level determination and feeds back and displays a corresponding retrieval result to the user, so that the user can directly select the corresponding retrieval result for fast input.

According to the embodiment of the invention, the object other than the intelligent information prompt assistant can be other users chatting with the user in a chat scene, can also be a chat robot talking with the user, and can also be an operation object of the user, such as an application program or a clipboard and the like. Accordingly, the natural interaction language may be chat information of the user with other users, dialog information of the user with the chat robot, text information selected by the user in an application, or text information copied to a clipboard by the user, etc.

According to an embodiment of the invention, the term "language category" may be any of the following language categories: chinese, English, Japanese, French, Hindi, German, Italian, Korean, Spanish, Portuguese, Russian, Belgium, Dutch, Arabic, Cyrillic, Greek, Indonesian, Malay, Philippine, Albarya, Basidion, Boganian, Bulgaria, Catalania, Crohn, Czerland, Denmark, Essah, Finnishia, Finnish, Callical, Hebrew, Hungarian, Iceland, Litotan, Malaya, Kazak, Gossypium, Laolaou, Latova, Marchant, Margarland, Margali, Marradland, Welan, Polish, Roman, Savian, Lavian, Lauraria, Royle, Laura, Russian, Switzerland Thailand, which, Vietnamese.

According to an embodiment of the present invention, the term "user representation" is used to describe a series of labels of a user, and the labels describing the user are formed by actively or passively collecting various data left by the user on the internet, and further analyzing and processing the data. "user representation" is used broadly to include at least one of personal information of a user, such as the user's age, gender, height, weight, nationality, native place, place of residence, place of employment, language of employment, idiomatic language, academic calendar, occupation, skill, hobbies, income status, etc.; information related to the user equipment, such as real-time location information of the user equipment, a model of the user equipment, a system language used by the user equipment, or other configuration information of the user equipment; user habits or preferences such as user eating habits, travel habits, sports preferences, travel preferences, and the like.

According to the embodiment of the invention, the retrieval result returned by the intelligent information prompt assistant system can comprise text information, picture information, audio information, video information, installation package or other format files and the like.

Referring to FIG. 1, the intelligent information prompting assistant system may include a processing device 110, a storage device 120, an input device 130, and a display device 140. Wherein the processor may be a central processing unit ("CPU") or a graphics processing unit ("GPU"), the processing device 110 may specifically include one or more printed circuit boards or micro-processing module chips that execute sequences of computer program instructions to perform various methods that will be explained in more detail below. In some embodiments, the processing device 110 may be configured to monitor a natural interaction language in which a user interacts with an object other than the intelligent information prompting assistant, determine whether information prompting service needs to be provided, retrieve the natural interaction language when service needs to be provided, and feed back a retrieval result to the user.

The storage device 120 may include one or more of random access memory ("RAM") and read only memory ("ROM"). The computer program instructions may be accessed and read from ROM or any other suitable memory location and loaded into RAM for execution by processing device 110. For example, storage device 120 may store one or more software applications. The software applications stored in the storage device 120 may include operating systems for general computer systems as well as for software-controlled devices. Furthermore, the storage device 120 may store the entire software application or only a portion of the software application that is executable by the processing device 110. For example, the storage device 120 may store information push software executable by the processing device 110 and execute an information push method.

According to certain embodiments, the storage device 120 may include a knowledge base adapted to store data matching services provided by the service modules. The number of the service modules can be multiple, and each service module can provide one service for the user. For example, the service module may be a weather service module adapted to provide weather information, such as providing a current temperature, air quality condition, cloudy/sunny status, whether there is rain or wind status, etc. for the user; it may also be an exchange rate service module adapted to provide exchange rate conversion information, such as to perform exchange rate conversion according to user requirements; the system can also be a recommendation service module suitable for providing recommendation information, such as a restaurant recommendation service module, a travel recommendation service module, a trip information recommendation service module, an exercise recommendation service module or other service modules for recommending preferential information; the service module can also be an intelligent reply service module and is suitable for providing contents needing reply for the user according to the information sent by the opposite user. The above description is only an example and is not a limitation of the present invention, and it can be understood by those skilled in the art that the service module may be any type of service module constructed according to the user requirement, and the corresponding service module may be added or subtracted according to the actual requirement.

According to the embodiment of the invention, the knowledge base can store data content corresponding to the service required to be provided by the service module, and the data is classified and stored according to the requirement of each service module to form the structured index knowledge base. Specifically, the data can be classified and stored by constructing different types of knowledge bases, for example, a weather knowledge base is constructed to store data content related to weather, an exchange rate knowledge base is constructed to store data content related to exchange rate, a restaurant recommendation knowledge base is constructed to store corresponding restaurant information, and the service module selects the corresponding knowledge base for retrieval during retrieval.

The data content stored in the knowledge base can be crawled in each website by the data crawling device, then the index is built according to the retrieval requirement, for example, for a weather service module, weather information needs to be retrieved according to time and position during retrieval, and the index can be built according to 'city' → 'time' in the weather knowledge base.

In addition, the data contents in the knowledge base can be classified and stored according to the language types, the data crawling device crawls data from corresponding websites according to the language types and stores the data into the knowledge base corresponding to the language types, for example, for a restaurant recommendation service module, the data crawling device can crawl the websites of different language types and store the data into the knowledge base corresponding to the restaurant recommendation service module according to the language types.

Further, the knowledge base can be a local knowledge base, can also be a cloud knowledge base, and can also be partially located locally and partially located in a cloud. In some embodiments, the storage device 120 also stores user historical chat information and data associated with a user representation, such as a user's historical input record, historical browsing record, a user's record of use of an application program, or historical operating record.

In some embodiments, the input device 130 and the display device 140 may be coupled to the processing device 110 by suitable interface circuitry. In some embodiments, the input device 130 may be a mouse, a keyboard, a touch pad, a touch screen, and may be adapted to obtain information input by a user, and operations performed by the user through the input device 130, such as a select operation, a copy operation, a cut operation, or other operations. In some embodiments, the input device 130 may also include a sensing input device 130, for example, the input device 130 may be a voice input device 130 and a voice parsing device, wherein a user performs a voice input through the voice input device 130, the voice parsing device detects that there is a voice input of the user, recognizes and parses the voice content input by the user, and the recognition result may be text information corresponding to the input voice. In some embodiments, the input device 130 may also include certain function keys by which a user may initiate certain processes performed by, or otherwise interact with, the intelligent information prompt assistant system.

Display device 140 may include one or more display screens that display text or graphics to a user. For example, the display device 140 may display a GUI.

In some embodiments, the intelligent kiosk system also includes a network interface 150, and the network interface 150 may provide a communication connection such that the intelligent kiosk system may connect to the cloud 160 through the network interface 150.

The intelligent information prompting assistant system is further explained in the following with the corresponding figures.

Referring to the block diagram of the intelligent information prompting assistant system shown in fig. 2, the processing device 110 further includes a monitoring device 111, a scenario distinguishing device 112, a service device 113, and a filtering and sorting device 114. The monitoring device 111 can monitor the natural interactive language of the user and transmit the natural interactive language to the situation judging device 112, the situation judging device 112 determines whether information prompt service needs to be provided after preprocessing and situation judging, when service needs to be provided, the natural interactive language is distributed to each service module of the service device 113, each service module determines whether the service of the service module needs to be provided through confidence judgment, when the service of the service module needs to be provided, the natural interactive language is retrieved, the filtering and sorting device 114 filters or sorts the retrieval result and then presents the retrieval result to the user, and the user can directly select the retrieval result to be displayed on a screen or look up the retrieval result. Therefore, the intelligent information prompt assistant can provide services for the user from the perspective of a third person and help the user to input information quickly.

Wherein, the monitoring device 111 is located in the mobile terminal; the scene discrimination device 112 may locally determine the natural interaction language, or may upload the natural interaction language to the cloud, and the cloud determines the natural interaction language to determine whether an information prompt service needs to be provided; the service device 113 includes at least one service module, which may be located locally or in the cloud; the mobile terminal may interact with the server via the network interface 150.

Wherein the monitoring device 111 is adapted to monitor a natural interaction language of the user interacting with an object other than the intelligent information prompting assistant, and according to an embodiment, the natural interaction language may be chat information occurring in a chat scene, such as dialog information of the user with other users, dialog information of the user with a chat robot, or information being input by the user.

For example, for the screen display interface shown in fig. 3a, the monitoring device 111 may monitor the information "weather" being input by the user, and for the screen display interface shown in fig. 3b, the monitoring device 111 may monitor the chat information "How's New York weather" sent by the opposite user.

According to another embodiment, the natural interaction language may also be information selected, copied or cut by the user in other scenarios, such as text information selected by the user in an application program, text information copied to a clipboard by the user, and when the user performs an operation of selecting text information, or a copy/paste operation, the monitoring device 111 obtains the corresponding natural interaction language.

In one embodiment, the monitoring device 111 may obtain the natural interactive language through a data interface provided by the relevant application program, for example, for the chat information in the WeChat, the monitoring device 111 may obtain the chat information sent by other users or the chat information sent by the user through the data interface provided by the WeChat; for another example, the monitoring device 111 obtains the information being input by the user through a data interface provided by the input method, or the input method directly transmits the information being input by the user to the monitoring device 111.

In another embodiment, the monitoring device 111 may also obtain the natural interactive language through a data interface provided by the system; in other embodiments, the monitoring device 111 may also perform real-time detection on the current screen display interface, identify icons and characters in the screen interface through a text-text identification technology, and then extract the required natural interactive language.

Further, the monitoring device 111 may monitor the update of the natural interactive language in real time, and transmit the updated natural interactive language to the situation discrimination device 112 in real time, and the situation discrimination device 112 may determine the natural interactive language in real time to determine whether to provide the information prompt service for the user at the current time. For example, when the user inputs a new character or a new word, the monitoring device 111 transmits the character or word input in real time to the scene discrimination device 112; for another example, when the user receives chat information transmitted from another user, the entire chat information is transmitted to the scene discrimination device 112. In other embodiments, the monitoring device 111 may also periodically acquire the natural interactive language at set time intervals, and transmit the natural interactive language to the scene discrimination device 112.

Next, the scene discrimination device 112 updates the received natural interactive language in real time, and determines whether to provide an information prompt service for the user according to the updated natural interactive language. The scene discrimination device 112 may determine whether to provide an information prompting service according to the current chat content, such as information currently input by the user, information currently sent by the user, or information currently sent by another chat object; the context determination device 112 may further determine whether the information prompting service needs to be provided according to the context information associated with the current chat content.

Specifically, the context determination device 112 may preprocess the natural interactive language, determine a language type to which the natural interactive language belongs, determine whether to provide an information prompt service for the natural interactive language, and send the preprocessed natural interactive language to the service device 113 when the service needs to be provided.

Referring to the structural block diagram of the exemplary scene discrimination device 112 shown in fig. 4, the scene discrimination device 112 specifically includes a preprocessing module and a discrimination module.

The preprocessing module is suitable for preprocessing the natural interactive language, determining the language type of the natural interactive language, and extracting features according to the natural interactive language to form structured data.

The preprocessing module can determine the language type of the natural interactive language through semantic analysis or a language recognition model, and further can comprehensively judge the language type of the natural interactive language by combining context information associated with the natural interactive language.

Further, the preprocessing module can also perform feature extraction according to common features required by more than two service modules and form structured data. For example, the weather service module and the exchange rate service module need to be retrieved according to the position information during retrieval, the position information can be used as the feature to be extracted, the weather service module and the exchange rate service module can directly acquire the position information, and therefore the feature can be extracted once to be used by a plurality of service modules, and the feature extraction efficiency is effectively improved.

According to an embodiment, the preprocessing module may extract features in the natural interaction language, for example, for the screen display interface shown in fig. 3b, when the monitoring device 111 monitors the natural interaction language "How's New York weather" sent by the opposite user, the preprocessing module extracts the location information "New York" and forms the structured data shown in table 1.

Position of	Natural interactive language
		New York	How’s New York weather

TABLE 1

According to another embodiment, the preprocessing module may further obtain user equipment information, such as IP information, GPS information, current time, model of the user equipment, and perform feature extraction on the user equipment information, for example, extract a city where the user is located according to the user IP information, and then form structured data according to the extracted features.

The judging module is suitable for judging whether information prompt service needs to be provided for the natural interactive language or not.

According to one embodiment, the judging module can select a corresponding classifier according to the language type of the natural interactive language to judge the natural interactive language and determine whether the information prompt service needs to be provided. The classifier may be a deep learning-based classifier or a linear model-based classifier. The classifier can be subjected to corpus training in advance, corpora can be user chat content or man-machine conversation content acquired from websites or databases corresponding to various language types, content information can be classified through the corpus training, which content needs to be provided with service and which content does not need to be provided with service can be determined, whether the current content needs to be provided with service can be judged when the classifier receives natural interactive language, when the content needs to be provided with service, the corresponding service module is used for retrieving and presenting retrieved information to a user, and therefore the intelligent information prompt assistant system can provide corresponding service for the user from the perspective of a third party and help the user to input information quickly.

Further, the judging module can also judge whether the local service module needs to be started to provide the information prompt service according to the natural interaction language monitored by the monitoring device 111, and when the service needs to be provided, the local service module is called to process the natural interaction language.

The local service module can be a service module with small calculation amount, so that the local service module is convenient for providing quick service for the user locally, and the pressure of the cloud service module is relieved to a certain extent. The local service module may be, for example, a computation service module that provides a result of mathematical operation, or may be an intelligent reply service module that provides reply information, which is described above only as an example and is not intended to limit the present invention.

Specifically, the judgment module can perform keyword recognition and semantic analysis on the natural interactive language according to the retrieval requirement of the local service module, and determine whether the local service module needs to be started to provide the information prompt service according to the analysis result, and when the service needs to be provided, the judgment module can retrieve the natural interactive language according to the local knowledge base, and can also call the local application program associated with the natural interactive language to process the natural interactive language.

For example, for the screen display interface shown in fig. 5a, the monitoring device 111 monitors that the information "It's 30 × 25" is being input by the user and sends the information "It's 30 × 25" to the determination module, and the determination module analyzes the information "It's 30 × 25" and determines that the current content is the service that the calculation service module needs to provide, then the calculation service module may invoke the local calculator to calculate 30 × 25, and display the calculated result "750" to the user, and the user may directly click to input the result into the input box.

For another example, for the screen display interface shown in fig. 5b, the monitoring device 111 monitors the information "How much I own you" sent by the user of the opposite party and sends the information "How much I own you" to the determination module, and the determination module analyzes the information "How much I own you" and then determines that the current content is the service that the intelligent reply service module needs to provide, so that the intelligent reply service module can match the corresponding reply information according to the local knowledge base and display the reply information "Costly | Not very outstanding" to the user, and the user can input the corresponding reply information into the input box by directly clicking the user.

According to an embodiment, when the scene discrimination device 112 receives the natural interaction language sent by the monitoring device 111, the natural interaction language may be preprocessed by the preprocessing module, and then the discrimination module may determine whether the information prompt service needs to be provided, and when the service needs to be provided, the preprocessed natural interaction language may be sent to the service device 113. The natural interactive language can determine the language type to which the natural interactive language belongs after being preprocessed, and the convenient distinguishing module can select a corresponding classifier according to the language type to distinguish whether the service needs to be provided.

According to another embodiment, when the scene discrimination device 112 receives the natural interaction language sent by the monitoring device 111, the discrimination module may first determine whether the information prompt service needs to be provided, and when the service needs to be provided, the preprocessing module may preprocess the natural interaction language, and then send the preprocessed natural interaction language to the service device 113. The judgment module can judge whether the service is needed to be provided or not through methods of semantic analysis, keyword extraction or classifier judgment and the like, and when the service is not needed to be provided, the natural interactive language is directly intercepted, so that the processing resources are effectively saved, and the processing efficiency is improved.

Next, when service needs to be provided, the scenario discrimination device 112 sends the preprocessed natural interaction language to each service module in the service device 113, and each service module further determines whether the service of the service module needs to be provided for the current natural interaction language, and when service needs to be provided, selects a corresponding knowledge base according to the language type to which the natural interaction language belongs to retrieve the natural interaction language.

For example, as for the screen display interface shown in fig. 5c, the context determination device 112 determines that the information prompt service needs to be provided after performing context determination on the information "How about Italian food" sent by the other user, then sends the natural interactive language to each service module in the service device 113, the other service modules determine that the service of the service module does not need to be provided after determination, and the restaurant recommendation service module determines that the restaurant service needs to be provided after determination, then searches in the corresponding knowledge base and provides the relevant restaurant information to the user.

For another example, for the screen display interface shown in fig. 5d, the scene discrimination device 112 determines that an information prompt service needs to be provided after performing scene discrimination on the information "Oh, I changed my mind" sent by another user, then sends the natural interaction language to each service module in the service device 113, the other service modules determine that the service of the service module does not need to be provided after the discrimination, and the intelligent reply module determines that a service that needs to provide reply information is needed after the discrimination, and then sends the reply information "Yeah, you? And "Not yet, why? And providing the user with corresponding prompt information to facilitate the user to quickly input the content to be replied.

Referring to the schematic structural diagram of the service module shown in fig. 6, the service module specifically includes a confidence level determining unit, a feature extracting unit, and an information feedback unit. Specifically, the confidence coefficient judgment unit performs confidence coefficient calculation on the natural interactive language to determine whether the current service module needs to provide the service of the service module, when the service needs to be provided, the feature extraction unit further performs feature extraction on the preprocessed natural interactive language according to the retrieval requirement of the current service module, then the information feedback unit selects a corresponding knowledge base according to the language type, performs retrieval in the corresponding knowledge base according to the extracted features, and feeds back the retrieval result to the display device.

According to one embodiment, the confidence degree judging unit may extract keywords from the natural interactive language according to preset keywords, determine confidence degrees of the extracted keywords, determine a comprehensive confidence degree of the natural interactive language according to the confidence degrees of the keywords, and determine whether the service of the service module needs to be provided according to the comprehensive confidence degree. The preset keywords correspond to the retrieval requirement of the service module, for example, the preset keywords corresponding to the weather service module may be "weather", or keywords associated with weather, such as "cold" and "hot". The confidence judging unit may add the confidences of the respective keywords to determine a comprehensive confidence.

According to another embodiment, the confidence judgment unit may further determine the confidence of the natural interaction language through a classifier, and determine whether the service of the service module needs to be provided according to the confidence. The classifier may be a deep learning-based classifier or a linear model-based classifier. Each service module can select corresponding corpora to train the classifier according to the service required by the service module.

The feature extraction unit can further extract features of the preprocessed natural interactive language according to the retrieval requirement of the current service module. For example, for the weather service module, the weather information needs to be retrieved according to time and location during retrieval, and the pre-processing module extracts the IP information of the user as a city during pre-processing, the feature extraction unit may further extract the time information. For another example, for the service module recommending restaurant information, when retrieving, a restaurant needs to be retrieved according to the user location information and information associated with the restaurant, and when preprocessing, the preprocessing module extracts the IP information of the user as a city, the feature extraction unit may further extract information associated with the restaurant in the natural interactive language, for example, extract "Italian" in "How about Italian food".

Further, the feature extraction unit may determine a corresponding feature extraction rule according to a language type to which the natural interactive language belongs, and perform feature extraction on the preprocessed natural interactive language according to the feature extraction rule. For example, english words are usually distinguished from one another by spaces, so that english words can be segmented by detecting spaces and corresponding punctuation marks, while chinese words are continuous from one another and are not suitable for segmentation by a method for detecting spaces, and segmentation is usually performed based on a dictionary for segmentation.

The information feedback unit can search in the corresponding knowledge base according to the characteristics extracted by the preprocessing module and the characteristic extraction unit and feed back the search result to the display equipment. Assuming that the preprocessing module obtains the feature 'Shanghai' when the user IP information is extracted, and the feature extracted by the feature extraction unit is 'Italian', the information feedback unit searches in the corresponding restaurant knowledge base according to the 'Shanghai' and the 'Italian', obtains an Italian restaurant in the Shanghai, and presents the Italian restaurant to the user through the display device for selection by the user.

Referring to fig. 3a, which shows a screen display interface, the retrieval result may be displayed in the form of bubbles; referring to fig. 3b, a screen display interface is shown, when the user clicks the bubble, the user can enter a detail page to display the detection result, and the user can directly click a sending option to send the weather information to the opposite user; referring to fig. 3c, which shows a screen display interface, the user may also input weather information into the current input box in a text form by performing corresponding operations, and the user may send the weather information to other users when clicking the send option.

Further, the service module further comprises a sorting unit adapted to sort the search results when the number of the search results obtained by the service module is more than two. According to an embodiment, the sorting unit may sort the search results according to a system parameter or a set sorting rule. Wherein, the system parameter can be user GPS information, for example; the set ordering rules may be, for example, ordering by sales, goodness, credit, distance, etc.

For example, referring to the screen display interface shown in fig. 5c, the information feedback unit retrieves italian restaurants located in shanghai, and the ranking unit can display the italian restaurants closer to the current position of the user to the user according to the GPS information of the user.

According to another embodiment, the sorting unit may further perform matching analysis on the search results according to the user image, adjust the display order of the search results so that the search results with high matching degree have higher display priority, and feed back the search results and the corresponding display priority to the display device. The user portrait can record the interests, hobbies, occupation, skills, habits, preferences and the like of the user, and the information feedback unit can adjust the display sequence of the retrieval results according to one label or a plurality of labels in the user portrait.

For example, the information feedback unit obtains an italian restaurant near the user position according to the user GPS information, and the ranking unit learns the user's habit to watch movies after having meals according to the user images, so that the ranking unit can increase the display priority of an italian restaurant closer to a movie theater in the current result, and adjust the display order of the italian restaurant according to the adjusted priority.

Next, when more than two service modules feed back the search result, the filtering and sorting device 114 may further filter the search result or adjust the display order of the search result.

According to an embodiment, the filtering and sorting device 114 may filter the search results of the service module or adjust the display order of the search results according to the confidence level calculated by the confidence level judging unit.

For example, if the weather service module and the intelligent reply service module give corresponding retrieval results at the same time, but the confidence of the weather service module is higher than that of the intelligent reply service module, the filtering and sorting device 114 may filter out the retrieval results of the intelligent reply service module and only present the retrieval results of the weather service module to the user; alternatively, the filtering and sorting means 114 increases the display priority of the retrieval result of the weather service module, thereby preferentially displaying the retrieval result of the weather service module to the user.

According to another embodiment, the filtering and sorting device 114 may filter the search results of the service module or adjust the display order of the search results according to the user image. For example, when the user receives chat information "How about Italian food" sent by the opposite user, the restaurant recommendation service module obtains an Italian restaurant near the user after retrieval, and the intelligent reply service module obtains reply information "Good idea | Sorry, I prefer …" after retrieval, the filtering and sorting device 114 learns that the user does not like eating Italian dishes through the user portrait, the filtering and sorting device 114 can filter out the retrieval result of the restaurant recommendation service module and only present the retrieval result of the intelligent reply service module to the user; alternatively, the filtering and sorting device 114 increases the display priority of the retrieval result of the intelligent reply service module, so that the retrieval result of the intelligent reply service module is preferentially displayed to the user.

Fig. 7 is a flowchart illustrating an information prompting method of an exemplary intelligent robot assistant system, where the method specifically includes:

and S110, monitoring natural interactive language of interaction between the user and objects except the intelligent information prompt assistant.

S120, preprocessing the natural interactive language, and determining the language type of the natural interactive language.

Further, the preprocessing module can also perform feature extraction according to the natural interactive language to form structured data. In particular, feature extraction may be performed based on common features required by more than two service modules. For example, when the restaurant recommendation service module and the exchange rate service module are searched according to the position information, the position information can be used as the feature to be extracted, for example, the current position information of the user is extracted according to the user IP information, or the real-time position information of the user is extracted according to the user GPS information, so that during subsequent processing, the restaurant recommendation service module and the exchange rate service module can directly obtain the position feature of the user, the feature can be extracted for use by a plurality of service modules only once, and the feature extraction efficiency is effectively improved.

Further, the preprocessing module may extract features in the natural interactive language, or extract features of other information associated with the user or the user equipment, such as user IP information, GPS information, current time, model of the user equipment, and the like

S130, judging whether information prompt service needs to be provided for the natural interactive language.

The corresponding classifier can be selected according to the language category to which the natural interactive language belongs to judge the natural interactive language and determine whether information prompt service needs to be provided. The analyzer can be trained by selecting corresponding corpora according to language types, the corpora can be user chat content or man-machine conversation content acquired from websites or databases corresponding to various language types, content information can be classified through the corpus training, which content needs to be provided with service and which content does not need to be provided with service can be determined, whether the current content needs to be provided with service can be judged when the classifier receives natural interaction language, when the service needs to be provided, the corresponding service module retrieves and presents retrieved information to a user, and therefore the intelligent information prompt assistant system can provide corresponding service for the user from the perspective of a third party and help the user to input information quickly.

In one embodiment, it may also be determined whether an information prompt service needs to be provided for the natural interaction language through S130, and when the service needs to be provided, the natural interaction language is preprocessed through S120. The method can judge whether the information prompt service needs to be provided or not through semantic analysis, keyword extraction or classifier judgment and the like, and directly intercept the natural interactive language when the service does not need to be provided, and follow-up steps do not need to be executed to process the natural interactive language, so that processing resources are effectively saved, and the processing efficiency is improved.

When information prompt service needs to be provided, the preprocessed natural interactive language is distributed to each service module in the service device, and each service module processes the natural interactive language respectively.

And S140, the service module selects a corresponding knowledge base according to the language type to retrieve the preprocessed natural interactive language.

Referring to the flow diagram of the service module shown in fig. 8 for processing the natural interaction language, the method specifically includes:

s141, performing confidence calculation on the natural interactive language, and determining whether the current service module needs to provide the service of the service module;

according to one embodiment, the method can extract keywords from the natural interactive language according to preset keywords, determine the confidence of each extracted keyword, determine the comprehensive confidence of the natural interactive language according to the confidence of each keyword, and further determine whether the service of the service module needs to be provided according to the comprehensive confidence.

According to another embodiment, the confidence of the natural interaction language can be determined through a classifier, and whether the service of the service module needs to be provided is determined according to the confidence.

Specifically, the confidence corresponding to the natural interaction language may be compared with a preset threshold, and when the confidence is greater than or equal to the preset threshold, it is determined that the service of the service module needs to be provided, and when the confidence is less than the preset threshold, it is determined that the service of the service module does not need to be provided.

S142, when the service of the service module needs to be provided, according to the retrieval requirement of the current service module, further performing feature extraction on the preprocessed natural interactive language;

for example, for the on-screen display interface shown in FIG. 9, the pre-processing module extracts the user IP information as "USA" when extracting the user IP information as "US" for the natural interaction language "I'll pay you 800 Euro", and the exchange rate service module can further extract "800" and "Euro" on this basis.

S143, selecting a corresponding knowledge base according to the language type, retrieving in the corresponding knowledge base according to the extracted characteristics, and feeding back a retrieval result to the display equipment;

for example, for the on-screen display interface shown in FIG. 9, the exchange rate service module may select the exchange rate knowledge base corresponding to English according to the current language category "English", search the exchange rate knowledge base according to the extracted features "800", "Euro", and the extracted user location "US", convert "800 Euro" into "976.15021 USD" according to the newest exchange rate of Euro dollars, and feed the converted results back to the display device.

And S150, sorting the search results.

When the number of the retrieval results obtained by one service module is more than two, the retrieval results can be ranked according to system parameters, set ranking rules or user images, so that the retrieval results with high matching degree have higher display priority. Wherein, the system parameter can be user GPS information, for example; the set ordering rules may be, for example, ordering by sales, goodness, credit, distance, etc. The user portrait can record the interests, hobbies, occupation, skills, habits, preferences and the like of the user, and the service module can adjust the display sequence of the retrieval results according to one label or a plurality of labels in the user portrait.

Further, when more than two service modules feed back the retrieval result, the retrieval result of the service modules can be filtered or the display sequence of the retrieval result can be adjusted according to the confidence degree corresponding to the natural interactive language or the user image.

For example, when the user receives chat information "How about Italian food" sent by the opposite user, the restaurant recommendation service module obtains an Italian restaurant near the user after retrieval, and the intelligent reply service module obtains reply information "Good idea | Sorry, I prefer …" after retrieval, the filtering and sorting device 114 learns that the user does not like eating Italian dishes through the user portrait, the filtering and sorting device 114 can filter out the retrieval result of the restaurant recommendation service module and only present the retrieval result of the intelligent reply service module to the user; alternatively, the filtering and sorting device 114 increases the display priority of the retrieval result of the intelligent reply service module, so that the retrieval result of the intelligent reply service module is preferentially displayed to the user.

And S160, displaying the retrieval result returned by the service module.

Referring to fig. 10, a flow diagram of an information prompting method of another exemplary intelligent robot assistant system is shown, where the method specifically includes:

s110, monitoring natural interactive language of interaction between a user and an object except the intelligent information prompt assistant;

s210, judging whether a local service module needs to be started to provide information prompt service or not;

specifically, the method can perform keyword recognition and semantic analysis on the natural interactive language, and determine whether the local service module needs to be started to provide the information prompt service according to the analysis result.

And S220, when the local service module needs to be started to provide the information prompt service, the local service module is called to process the natural interactive language.

According to one embodiment, natural interactive language may be retrieved according to a local knowledge base; according to another embodiment, a native application associated with the natural interaction language may also be invoked to process the natural interaction language.

And S230, when the information prompt service provided by the local service module is not required to be started, sending the natural interactive language to the cloud to judge whether the information prompt service is required to be provided for the natural interactive language.

For example, the natural interaction language may be processed at the cloud by the method flow of S120-150 as shown in fig. 7.

It is to be noted that the foregoing is only illustrative of the preferred embodiments of the present invention and the technical principles employed. It will be understood by those skilled in the art that the present invention is not limited to the particular embodiments described herein, but is capable of various obvious changes, rearrangements and substitutions as will now become apparent to those skilled in the art without departing from the scope of the invention. Therefore, although the present invention has been described in greater detail by the above embodiments, the present invention is not limited to the above embodiments, and may include other equivalent embodiments without departing from the spirit of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. An intelligent information prompting assistant system, comprising:

the monitoring device is suitable for monitoring natural interactive language of interaction between a user and an object except the intelligent information prompt assistant;

the scene discrimination device is suitable for preprocessing the natural interactive language, determining the language type of the natural interactive language, judging whether information prompt service needs to be provided for the natural interactive language or not, and sending the preprocessed natural interactive language to the service device when the service needs to be provided;

the service device comprises at least one service module, and the service module is suitable for selecting a corresponding knowledge base according to the language type to retrieve the preprocessed natural interactive language;

the display device is suitable for displaying the retrieval result returned by the service module;

and the knowledge base is suitable for storing data matched with the service provided by the service module.

2. The system according to claim 1, wherein the scene discrimination device specifically includes:

the preprocessing module is suitable for preprocessing the natural interactive language, determining the language type of the natural interactive language, and extracting features according to the natural interactive language to form structured data;

and the judging module is suitable for judging whether the information prompt service needs to be provided for the natural interactive language.

3. The system of claim 2, wherein the discrimination module is further adapted to:

and judging whether a local service module needs to be started to provide information prompt service according to the natural interactive language monitored by the monitoring device, and calling the local service module to process the natural interactive language when the service needs to be provided.

4. The system of claim 3, wherein invoking the local services module to process the natural interaction language comprises:

retrieving the natural interactive language according to a local knowledge base; or

And calling a local application program associated with the natural interaction language to process the natural interaction language.

5. The system of claim 3, further comprising:

when the information prompt service provided by the local service module is not needed to be started, the judgment module sends the natural interaction language to the cloud to judge whether the information prompt service is needed to be provided for the natural interaction language.

6. The system of claim 2, wherein the pre-processing module performs feature extraction based on common features required by two or more service modules and forms the structured data.

7. The system of claim 2, wherein the preprocessing module preprocesses the natural interactive language, the determining module determines whether an information prompt service needs to be provided, and the preprocessing module sends the natural interactive language to the service device when the service needs to be provided.

8. The system of claim 2, wherein the determining module determines the natural interactive language monitored by the monitoring device to determine whether an information prompt service needs to be provided, and when the service needs to be provided, sends the natural interactive language to the preprocessing module for preprocessing, and sends the preprocessed natural interactive language to the service device.

9. The system according to claim 1, wherein the service module specifically comprises:

the confidence judgment unit is suitable for carrying out confidence calculation on the natural interactive language and determining whether the current service module needs to provide the service of the service module;

the feature extraction unit is suitable for further extracting features of the preprocessed natural interactive language according to the retrieval requirement of the current service module;

and the information feedback unit is suitable for selecting the corresponding knowledge base according to the language type, searching in the corresponding knowledge base according to the extracted characteristics and feeding back the search result to the display equipment.

10. The system of claim 9, wherein the service module further comprises:

and the sequencing unit is suitable for performing matching analysis on the retrieval results according to the user images when the retrieval results obtained by the service module are more than two, adjusting the display sequence of the retrieval results, enabling the retrieval results with high matching degree to have higher display priority, and feeding back the retrieval results and the corresponding display priority to the display equipment.

11. The system of claim 1, further comprising:

the filtering and sequencing device is suitable for filtering the retrieval result according to the user image when more than two service modules feed back the retrieval result; or

The display order of the search results is adjusted according to the user image.

12. The system of claim 1, wherein the knowledge base classifies and stores data according to the requirements of each service module to form a structured index knowledge base.

13. The system of claim 1, wherein the natural interaction language comprises chat information for a user to interact with other users via a chat application.

14. The system of claim 1, wherein the natural interaction language further comprises at least one of:

chat information used for interacting with other chat robots;

the text information copied to the clipboard by the user;

and the user selects the text information in the application program.

15. An information presentation method for an intelligent robot assistant system, comprising:

monitoring natural interactive language of interaction between a user and objects except the intelligent information prompting assistant;

preprocessing the natural interactive language, and determining a language type to which the natural interactive language belongs;

judging whether information prompt service needs to be provided for the natural interactive language, and sending the preprocessed natural interactive language to a service device when the service needs to be provided, wherein the service device comprises at least one service module;

the service module selects a corresponding knowledge base according to the language type to retrieve the preprocessed natural interactive language;

and displaying the retrieval result returned by the service module.

16. The method of claim 15, wherein pre-processing further comprises:

and performing feature extraction according to the natural interactive language to form structured data.

17. The method of claim 16, wherein pre-processing further comprises:

feature extraction is performed according to common features required by more than two service modules, and structured data is formed.

18. The method of claim 15, further comprising, prior to preprocessing:

and judging whether the local service module needs to be started to provide the information prompt service according to the natural interactive language, and calling the local service module to process the natural interactive language when the service needs to be provided.

19. The method of claim 18, wherein invoking the local services module to process the natural interaction language comprises:

20. The method of claim 18, further comprising:

and when the information prompt service provided by the local service module is not required to be started, the natural interactive language is sent to the cloud to judge whether the information prompt service is required to be provided for the natural interactive language.

21. The method of claim 15, wherein selecting the corresponding knowledge base according to the language category to retrieve the preprocessed natural interactive language specifically comprises:

performing confidence calculation on the natural interactive language, and determining whether the current service module needs to provide the service of the service module;

when the service of the service module needs to be provided, further extracting the characteristics of the preprocessed natural interactive language according to the retrieval requirement of the current service module;

and selecting a corresponding knowledge base according to the language type, retrieving in the corresponding knowledge base according to the extracted characteristics, and feeding back a retrieval result to the display equipment.

22. The method of claim 5, further comprising:

when the number of the retrieval results obtained by the service module is more than two, the retrieval results are subjected to matching analysis according to the user image, the display sequence of the retrieval results is adjusted, the retrieval results with high matching degree have higher display priority, and the retrieval results and the corresponding display priority are fed back to the display equipment.

23. The method of claim 15, further comprising:

when more than two service modules feed back the retrieval result, filtering the retrieval result according to the user image; or

24. A terminal device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the program executed by the processor comprises the method according to any of claims 15-23.