CN111512617A

CN111512617A - Device and method for recommending contact information

Info

Publication number: CN111512617A
Application number: CN201880082700.6A
Authority: CN
Inventors: 黄陈煐
Original assignee: Samsung Electronics Co Ltd
Current assignee: Samsung Electronics Co Ltd
Priority date: 2017-12-22
Filing date: 2018-12-21
Publication date: 2020-08-07
Anticipated expiration: 2038-12-21
Also published as: EP3652925A1; CN111512617B; CN113746978A; EP3652925A4; KR102628042B1; KR20190076870A

Abstract

An apparatus is provided. The apparatus includes a processor and a memory configured to store instructions executable by the processor. The processor is configured to execute instructions to extract context information from the displayed data based on an application being executed by the apparatus, identify an identifier from the context information, search for at least one recommended contact related to the identifier based on the identifier and a relationship graph, identify a priority of the at least one recommended contact, and control to display the at least one recommended contact according to the priority, wherein the relationship graph is obtained by inputting information about communications between a plurality of users to a first training model for determining associations between the plurality of users.

Description

Device and method for recommending contact information

Technical Field

The present disclosure relates to a method and apparatus for recommending contact information. More particularly, the present disclosure relates to a method and apparatus for recommending contact information using current context information of a user and a contact method.

Background

Unlike existing rule-based intelligence systems, Artificial Intelligence (AI) systems are a computer system that can achieve a level of human intelligence and enable machines to self-learn, self-determine, and become more intelligent. The more AI systems that are used, the higher the recognition (recognition) rate, and thus the more accurate understanding of the user's preferences can be achieved. Thus, existing rule-based intelligent systems are gradually being replaced by deep learning based AI systems.

The AI technique is composed of machine learning (deep learning) and elementary techniques using machine learning (elementary techniques).

Machine learning is an algorithmic technique for self-classifying/learning features of input data. The basic technique is a technique of simulating a cognitive function, a determination function, and the like of the human brain using a machine learning algorithm such as deep learning, and can be classified into a plurality of technical fields including, for example, language understanding, visual understanding, inference/prediction, knowledge representation, operation control, and the like.

Various fields to which the AI technique is applicable will be described below. Language understanding is a technique to recognize and apply/process human language/characters, including natural language processing, machine translation, dialog systems, question answering, speech recognition/synthesis, and the like. Visual compression is a process of identifying and processing objects in terms of human perspectives, including object recognition, object tracking, video searching, human recognition, scene understanding, knowledge of space, video enhancement, and the like. Inference/prediction is a technique for judging and logically inferring information and making predictions, including knowledge/probabilistic-based inference, optimization predictions, preference-based planning, recommendations, and so forth. Knowledge representation is a technique for automatically processing human experience information based on knowledge data, including knowledge building (data creation/classification), knowledge management (data utilization), and the like. The operation control is a technique for controlling the automatic driving of the vehicle and the activity of the robot, and includes motion control (navigation, collision, driving), steering control (behavior control), and the like.

With the development of communication technology, various functions of communicating with a person have been provided to a user terminal, and thus not only the number of contacts for making a voice call, sending a text message, and the like, but also the number of contacts provided for communication through a messenger, an e-mail, a Social Network Service (SNS), and the like have increased.

The above information is provided merely as background information to aid in understanding the present disclosure. No determination is made as to whether any of the above would be applicable to the prior art in connection with the present disclosure.

Disclosure of Invention

Technical problem

It is difficult to remember a large number of contacts for various types of communication channels. For the convenience of the user, each communication channel program provides a function of recommending contacts in various ways. However, in most existing methods of recommending contacts, contacts are simply recommended based only on the number of calls and the duration of each call. Therefore, there is an increasing need to subdivide and recommend contacts based on the user's context.

Solution scheme

The present invention provides an apparatus comprising a processor and a memory configured to store instructions executable by the processor. The processor is configured to execute instructions to extract context information from the displayed data based on an application being executed by the apparatus, identify an identifier from the context information, search for at least one recommended contact related to the identifier based on the identifier and a relationship graph, identify a priority of the at least one recommended contact, and control to display the at least one recommended contact according to the priority, wherein the relationship graph is obtained by inputting information about communications between a plurality of users to a first training model for determining associations between the plurality of users.

Advantageous effects

The present disclosure enables providing a method and apparatus for recommending contacts based on context information and a relationship with a user.

Drawings

The above and other aspects, features and advantages of certain embodiments of the present disclosure will become more apparent from the following description taken in conjunction with the accompanying drawings in which:

FIG. 1 is an overview diagram of a method of recommending contacts performed by an apparatus according to an embodiment of the present disclosure;

FIG. 2 is a flow chart of a method of operating a device according to an embodiment of the present disclosure;

fig. 3 is a diagram illustrating a method of recommending contacts by using an identifier, the method being performed by an apparatus, according to an embodiment of the present disclosure;

FIG. 4 is a schematic diagram illustrating a method of creating a relationship graph based on affinity between contacts according to an embodiment of the present disclosure;

FIG. 5 is a schematic diagram illustrating a method of estimating recommended contacts from a user's mail by using a user relationship graphic, the method being performed by an apparatus, according to an embodiment of the present disclosure;

FIG. 6 is a schematic diagram illustrating a method of displaying a contact method linked to a selected recommended contact, the method being performed by an apparatus, according to an embodiment of the present disclosure;

FIG. 7 is a schematic diagram illustrating a method of displaying another contact method linked to a recommended contact when there is no connection with the recommended contact according to an embodiment of the present disclosure;

FIG. 8 is a schematic diagram illustrating a method of recommending another recommended contact according to an embodiment of the present disclosure;

FIG. 9 is a diagram illustrating a method of estimating recommended contacts based on schedule information and location information according to an embodiment of the present disclosure;

fig. 10 is a diagram illustrating a method of creating an address book after obtaining a new contact, the method being performed by an apparatus, according to an embodiment of the present disclosure;

FIG. 11 is a diagram illustrating a method of updating an address book based on user relationship graphics, the method being performed by an apparatus, according to an embodiment of the present disclosure;

FIG. 12 is a schematic diagram illustrating a message recommendation based on schedule information and group information, the recommendation being performed by a device, according to an embodiment of the present disclosure;

FIG. 13 is a block diagram of an apparatus according to an embodiment of the present disclosure;

FIG. 14 is a detailed block diagram of an apparatus according to an embodiment of the present disclosure; and

FIG. 15 is a block diagram of a processor according to an embodiment of the disclosure.

Throughout the drawings, the same reference numerals will be understood to refer to the same parts, assemblies and structures.

Detailed Description

Aspects of the present disclosure are directed to solving at least the above problems and/or disadvantages and to providing at least the advantages described below. Accordingly, one aspect of the present disclosure is to provide a method and apparatus for recommending contacts based on context information and a relationship with a user of the user.

Additional aspects will be set forth in part in the description which follows and, in part, will be obvious from the description, or may be learned by practice of the presented embodiments.

According to one aspect of the present disclosure, an apparatus is provided. The apparatus includes a processor and a memory configured to store instructions executable by the processor. The processor is configured to execute instructions to: the method includes extracting context information from displayed data based on an application being executed by an apparatus, identifying an identifier from the context information, searching for at least one recommended contact related to the identifier based on the identifier and a relationship graph, identifying a priority of the at least one recommended contact, and controlling to display the at least one recommended contact according to the priority, wherein the relationship graph is obtained by inputting information regarding communication between a plurality of users to a first training model for determining an association between the plurality of users.

According to another aspect of the present disclosure, a method is provided. The method comprises the following steps: the method includes extracting context information from displayed data based on an application being executed by an apparatus, identifying an identifier from the context information, searching for at least one recommended contact related to the identifier based on the identifier and a relationship graph, determining a priority of the at least one recommended contact, and displaying the at least one recommended contact according to the priority, wherein the relationship graph is obtained by inputting information regarding communications between a plurality of users to a training model for determining associations between the plurality of users.

According to another aspect of the present disclosure, a non-transitory computer-readable recording medium is provided. The non-transitory computer-readable recording medium has a program recorded thereon, which when executed by a computer performs the above-described method.

Other aspects, advantages, and salient features of the disclosure will become apparent to those skilled in the art from the following detailed description, which, taken in conjunction with the annexed drawings, discloses various embodiments of the disclosure.

Modes for carrying out the invention

The following description with reference to the accompanying drawings is provided to assist in a comprehensive understanding of various embodiments of the disclosure as defined by the claims and their equivalents. The following description includes various specific details to aid understanding, but these are to be considered exemplary only. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the various embodiments described herein can be made without departing from the scope and spirit of the present disclosure. In addition, descriptions of well-known functions and constructions may be omitted for clarity and conciseness.

The terms and words used in the following description and claims are not limited to the written meaning, but are used only by the inventors to enable a clear and consistent understanding of the disclosure. Accordingly, it will be apparent to those skilled in the art that the following descriptions of the various embodiments of the present disclosure are provided for illustration only and not for the purpose of limiting the disclosure as defined by the appended claims and their equivalents.

It is understood that the singular forms "a", "an" and "the" include plural referents unless the context clearly dictates otherwise. Thus, for example, reference to "a component surface" includes reference to one or more of such surfaces.

Throughout the disclosure, it will be understood that when an element is referred to as being "connected to" another element, it can be directly connected to or electrically connected to the other element with intervening elements therebetween. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated elements, but do not preclude the presence or addition of one or more other elements. As used herein, the term "and/or" includes any and all combinations of one or more of the associated listed items. When an expression such as "at least one" precedes a list of elements, the entire list of elements is modified without modifying individual elements of the list.

Embodiments will be described in more detail below with reference to the accompanying drawings.

Fig. 1 is an overview diagram of a method of recommending contacts, performed by an apparatus, according to an embodiment of the present disclosure.

Referring to fig. 1, in one embodiment, the apparatus 10 may identify at least one identifier, such as a first identifier 101, a second identifier 102, and a third identifier 103, from context information of an application being executed in the apparatus 10.

In one embodiment, the identifier may be a text keyword or a facial feature point identified by facial recognition. The device 10 may identify at least one identifier and a group identifier that are related to each other.

In one embodiment, the apparatus 10 may extract words using a specific frequency or higher in the context information as identifiers by using a language processing algorithm. In this case, the identifier may be a keyword.

In one embodiment, the apparatus 10 may obtain the keywords by maintaining a mapping between two or more dictionaries such that a word or phrase of one dictionary maps to another word or phrase of another dictionary. In one embodiment, when there is an association between a word or phrase in one dictionary and a word or phrase in another dictionary, the apparatus 10 may obtain keywords that are interchangeable with each other by using the mapping. The association may represent a method associated with a pair of nodes corresponding to a pair of data. In various embodiments, associations may be represented by predicates (predicates) having a pair of data or nodes as subject and object, respectively. In one embodiment, the association between two or more dictionaries may include, but is not limited to, the same relationship or an inclusion relationship. For example, unless the context requires otherwise, the words "korea" and "KR" having the same relationship are interchangeable with one another. Unless the context requires otherwise, the words "sports" and "basketball" are interchangeable as inclusive. The dictionary may be a language dictionary, a search context dictionary, an AdWords dictionary, a user intent map, or a user defined dictionary. The language processing algorithm may be, for example, a neural language programming algorithm, a natural language processing algorithm, or the like, or may include various algorithms or combinations thereof for analyzing and processing human language. In one embodiment, the apparatus 10 may extract a face detected at a specific frequency or higher from the context information as an identifier by using a facial recognition algorithm. In this case, the identifier may be a facial feature point.

The apparatus 10 may detect feature points on a face region by using coordinate information corresponding to the face.

In one embodiment, the apparatus 10 may obtain context information for an application currently being executed. In one embodiment, the application may include a mail, a Social Network Service (SNS), a text, an album, a diary, news, and the like, but the type of the application is not limited thereto. In one embodiment, the context information may include photos, pictures, text, moving pictures, animations, etc. included in the application. Optionally, in one embodiment, the context information may include a current time, a runtime of the application, properties of the application, a type of the application, and the like.

In one embodiment, the apparatus 10 may obtain the context information from the data displayed thereon by executing an application in the apparatus 10. In one embodiment, the apparatus 10 may obtain contextual information based on the text displayed thereon. For example, the text displayed on the apparatus 10 may include at least one of an email header, an email body, a sender, and a recommender (referrer) displayed thereon by executing an email application. In one embodiment, the apparatus 10 may obtain the context information by using different weights according to the fields including the text. For example, the apparatus 10 may obtain the context information by assigning the maximum weight to the text included in the title of the email and the minimum weight to the text included in the recommender. Alternatively, the apparatus 10 may obtain the context information by classifying the text into different categories according to the fields including the text. For example, the apparatus 10 may obtain the context information by classifying the text included in the header of the electronic mail into a "destination" category and classifying the text included in the sender into a "person" category.

Alternatively, the apparatus 10 may obtain the context information based on the image displayed thereon. For example, the image displayed on the apparatus 10 may include at least one of a moving image, a still image, and a representative image displayed for a certain time by executing a moving picture application. In one embodiment, the apparatus 10 may use different weights to obtain the context information according to the fields comprising the image. Alternatively, the apparatus 10 may obtain the context information by classifying the text into different categories according to the fields including the image. In one embodiment, the apparatus 10 may analyze the context information, extract at least one identifier, e.g., at least one keyword and at least one facial feature point, and identify a trend of the context information by using the extracted identifier. For example, when the first identifier 101 is "mother's face", the second identifier 102 is "family travel", and the third identifier 103 is "joy", the apparatus 10 may recognize that the trend of the currently performed context information is "pleasant family travel". In one embodiment, the apparatus 10 may process the plurality of identifiers into text representing a time series of operations based on the plurality of identifiers to determine a trend of the context information. For example, the apparatus 10 may generate text such as "the user has searched for famous places of travel in canada", "the user has searched for restaurant information in canada", and "the user goes to and John and Tom play basketball", based on the plurality of identifiers, and identify trends in the context information based on the generated text.

In one embodiment, the apparatus 10 may search for a recommended contact associated with the identifier. The apparatus 10 may search for a plurality of recommended contacts associated with a plurality of identifiers. The apparatus 10 may search for a recommended contact associated with the at least one identifier using a training model based on deep neural network techniques.

The apparatus 10 may search for a recommended contact associated with one or a combination of identifiers. For example, when the first identifier 101 is "mother's face", the second identifier 102 is "family travel", and the third identifier 103 is "joy", the apparatus 10 may search for family contacts as recommended contacts.

In one embodiment, the apparatus 10 may identify a priority for searching at least one recommended contact. In one embodiment, the apparatus 10 may identify a priority of the recommended contact in consideration of a contact frequency, a contact time, a contact method, a contact period (period), and the like with the recommended contact.

In one embodiment, the apparatus 10 may display the recommended contacts on a portion of the display. In one embodiment, the device 10 may display the recommended contacts in a pop-up window, on a separate window, or in the form of a message.

In one embodiment, the apparatus 10 may display the recommended contacts in a pop-up window 104 at the lower end of the screen. In one embodiment, the pop-up window 104 may include Identification (ID) information 106 of the recommended contact, e.g., the name of the stored recommended contact. In one embodiment, the pop-up window 104 may include an icon 107 for recommending a contact method for the contact. In one embodiment, the detailed information 105 of the recommended contacts may be displayed in a pop-up window 104. The detailed information 105 of the recommended contact may include a phone number when the contact method is a phone call, and may include ID information when the contact method is an SNS.

Fig. 2 is a flow chart of a method of operating a device according to an embodiment of the present disclosure.

Referring to fig. 2, in operation 201, in one embodiment, the apparatus 10 may identify an identifier from context information of an executing application. In one embodiment, the apparatus 10 may identify the at least one identifier in the context information by using a language processing algorithm or a facial recognition algorithm. In one embodiment, the identifier may be identified using words, synonyms, similar words, etc. that are detected at a certain frequency or higher.

In operation 202, in one embodiment, the apparatus 10 may search for at least one contact associated with the identifier. In one embodiment, the apparatus 10 may obtain an Artificial Intelligence (AI) training model based on a learning (graphical learning) result of a relationship between an identifier and a plurality of pieces of contact information using information on a history of the identifier used for a certain period of time.

For example, the apparatus 10 may process the plurality of identifiers into text representing a time series of operations. For example, the apparatus 10 may create text such as "the user has searched for famous places of travel in canada", "the user has searched for restaurant information in canada", and "the user goes and John and Tom play basketball", based on the plurality of identifiers.

In one embodiment, the apparatus 10 may learn relationships between a user and other users (e.g., contact owners) or between another user and other users by using information about communications between the user and the other users. The information about the communication may include frequency of contact, time of contact, location of contact, affinity of contact, common interests, affinity, etc. between the plurality of users. The apparatus 10 may be trained by inputting the above-described communication information into a first training model for determining an association between a plurality of users, and may generate a relationship graph based on the training result.

For example, the apparatus 10 may generate a relationship graph that is trained to match other users associated with travel or foreign countries that contact the user very frequently with the user in a "travel" relationship. Alternatively, the apparatus 10 may generate a relationship graph that is trained to match other users who contact the user very frequently who are performing the same workout to the user or users in the "workout" relationship. The relationship graph will be described in more detail below with reference to fig. 4.

In one embodiment, the apparatus 10 may identify an association between the identifier and the relationship graph by using an AI training model and search for a recommended contact based on the identified association. In one embodiment, the AI training model may be learned to recognize associations between identifiers and relationship graphs using text processed using the time series operations described above.

For example, when extracting "canada" as the identifier, the apparatus 10 may identify contacts of other users that match the user in the "travel" relationship in the relationship graph as recommended contacts by using an AI training model trained based on information about the history of use of "canada" with "travel".

When "John's face" is extracted as the identifier, the apparatus 10 may identify, as the recommended contact, a contact of Tom that matches John in the friendship relationship in the relationship graph by using an AI training model trained based on information about history of use of "John's face" with "Tom's face". However, the above-described case is merely an example, and the AI training model trained based on the historical information and the relationship graph according to the present disclosure are not limited thereto.

In one embodiment, the apparatus 10 may search for at least one recommended contact by using the obtained AI training model.

In operation 203, in one embodiment, the apparatus 10 may identify a priority of the searched at least one recommended contact. In one embodiment, the apparatus 10 may identify the priority of the at least one recommended contact based on historical information such as the number of contacts with the at least one recommended contact, a contact method, a contact period, and a relationship with an identifier and current information such as a current time, a current location, and information about an application currently being used.

In operation 204, in one embodiment, the apparatus 10 displays at least one recommended contact according to the priority of the recommended contact. In one embodiment, the apparatus 10 may display a plurality of recommended contacts simultaneously or sequentially. In one embodiment, the apparatus 10 may simultaneously display at least one connection method linked to at least one recommended contact or may display one of the at least one connection methods at a time.

Fig. 3 is a schematic diagram illustrating a method of recommending a contact by using an identifier, the method being performed by an apparatus, according to an embodiment of the present disclosure.

Referring to fig. 3, in one embodiment, the apparatus 10 may identify a first identifier 101, a second identifier 102, and a third identifier 103 in the context information.

In one embodiment, the apparatus 10 may search for at least one recommended contact by using an AI training model obtained based on a result of learning the relationship between the identifier and the pieces of contact information (graph learning).

In one embodiment, the device 10 may assign weights to the contact information used based on the application, duration, time, frequency, and location of the use identifier. Further, the apparatus 10 may identify at least one recommended contact based on the assigned weight.

In one embodiment, the apparatus 10 may collect a history of usage of at least one personal contact function (such as calls, text messages, messenger, email, SNS, etc.). In one embodiment, the usage history may include the date and time when each function was used to make the contact. In addition, the usage history may include a call time when the call function is provided, and may include contact details for using the remaining function when the call function is not provided.

In one embodiment, the apparatus 10 analyzes the collected usage history of each function according to at least one criterion of duration, time and frequency to assign a weight to the used contact. For example, a contact that was frequently used in the last few days may be selected by assigning a weight of 2.0 to a contact that was used within the last two days, or a contact that was frequently used in a time zone in which the user was primarily communicating with an intimate contact may be selected by assigning a weight of 1.5 to a contact that was used from 6 pm to 12 pm.

In one embodiment, the apparatus 10 may assign a weight to a contact that has exchanged words or emoticons expressing a predetermined emotion in the usage history of a personal contact function other than the call function. For example, a contact with a high affinity may be selected by assigning a weight of 2.0 to an email contact history that includes the word "love" or its emoticon. Conversely, a contact related to a loan advertisement or the like may be selected to have a lower priority by assigning a weight of 0.1 to the email contact history including the word "loan".

In one embodiment, the apparatus 10 may select at least one contact as the recommended contact by identifying a priority of the contacts for each function.

In one embodiment, when one of the recommended contacts corresponding to each function displayed on the screen is selected, the apparatus 10 may activate a contact method linked to the selected contact. In addition, the apparatus 10 may assign a weight to the recommended contact selected by the user, so that the frequently selected contact may be recommended more frequently. In one embodiment, the apparatus 10 may assign a weight to contacts that have been selected more than or equal to a predetermined number of times. Thus, the apparatus 10 can achieve an effect of activating the use of the recommended contact more.

In one embodiment, the apparatus 10 may identify a recommended contact, including the recommender 302 and the contact method 303. In one embodiment, the recommender 302 may be identification information, such as a person or company highly related to at least one of the first identifier 101, the second identifier 102, and the third identifier 103. In one embodiment, the apparatus 10 may identify an identification name for identifying a recommended contact received from an internal repository, cloud, or server and associated with a plurality of identifiers. In one embodiment, the identification name may include a name stored in the device 10, a company name searched through a network, and identification information on the recommender 302 searched through an SNS.

In one embodiment, the apparatus 10 may search for at least one contact corresponding to the recommender 302. In one embodiment, the contacts corresponding to the recommender 302 may include an SNS address, an available phone number, a mail address, an SNS ID, and the like. In one embodiment, the apparatus 10 may identify the priority of the recommender 302 by reflecting information detected in association with at least one identifier. For example, a contact of the family member "mother" may be given priority when at least one identifier (e.g., the first identifier 101, the second identifier 102, and the third identifier 103) is associated with a family trip.

In one embodiment, the apparatus 10 may identify a priority corresponding to the contact method 303 of the recommender 302. In one embodiment, the apparatus 10 may identify the priority of the contact method 303 by reflecting information detected in association with at least one identifier. In one embodiment, the apparatus 10 may identify a priority of the contact method 303 corresponding to the recommender 302 based on weights assigned according to frequently used contact methods, most recently used contact methods, contact locations, and the like. For example, when a phone call is often used as one of the recommenders 302 and its priority 301 is "mother" for the number 1 of family trips, it may be determined that the phone call is the first level (rank) contact method 304 linked to mother. Similarly, the text message may be determined as a second level contact method 305, and the SNS may be determined as a third level contact method 306.

FIG. 4 is a schematic diagram illustrating a method of creating a relationship graph from affinity between contacts according to an embodiment of the disclosure.

In one embodiment, the apparatus 10 may learn relationships between the user and other users based on information related to communications between the user and other users. In one embodiment, the apparatus 10 may learn the relationship by using a frequency of contact with other users, a time of contact, a location of contact, an affinity of contact, a common interest, an affinity, etc., and may create a relationship graph based on a result of learning the relationship.

In one embodiment, the apparatus 10 may reflect untagged information such as information newly disclosed through SNS, mail, phone call, etc. to the initial graph based on tagged information such as names of other users, contacts, groups to which the other users belong, frequency of contact, etc. stored in advance.

Referring to fig. 4, the circle nodes represent people. The location and color of the nodes may be data fields indicating similarity, interest, intimacy, etc. between the nodes. In one embodiment, a node may include a plurality of patterns, each pattern occupying an area. The shape of the pattern may represent different data fields, and the area occupied by the pattern in the node may represent the relative ratio between the values of the data fields. In one embodiment, for example, in the first graph of FIG. 4, B1 and B2 have a common interest in a common data field, namely, "Canada". In the relational graph, R1 is interested in the data field, i.e., "travel".

In one embodiment, each line connecting nodes may represent a relationship between users corresponding to the nodes. The thickness (thickness) of each line connecting nodes may represent the strength of the relationship between the nodes. The strength of the relationship may include affinity, similarity, whether each node belongs to a common group, and the like. For example, in the first group of fig. 4, the strength of the relationship between B2 and G3 may be greater than the strength of the relationship between G3 and G2. In the graph of fig. 4, the thickness of each line connecting nodes may be a value indicating the strength of a relationship normalized based on data fields included in the respective nodes. However, the thickness of each line is merely an example, and in the graph of fig. 4, each line connecting nodes may be provided for one of the data fields. For example, B2 and G3 may be connected by a line having a first thickness based on the "travel" data field and may be connected by a line having a second thickness based on the "canadian" data field. Alternatively, each line connecting nodes may represent a field indicating an association as well as a value. For example, fields indicating associations represented by lines connecting B2 and G3 may include the same relationship sameAs, membership relationship type, containment relationship subbackof, and so on.

In one embodiment, the device 10 may create the initial graphic using information obtained based on predefined people and communication information. In the first graph of fig. 4, G1, G2, and G3 have not been learned.

With the first training, the device 10 may train G2 based on the content of the conversation, message, or mail between B1 and G2. Referring to the second graph of fig. 4, it is noted that the similarity of G2 related to canada increases due to the effect of B1. Meanwhile, G3 may be trained based on the relationship between B2 and R1. Since the strength of the relationship between G3 and B2 is greater than the strength of the relationship between G3 and R1, G3 can be trained by assigning higher weight to canada than to travel.

Referring to the third graph of fig. 4, through the second training, the apparatus 10 may learn a relationship graph based on information related to communication of B1, B2, G1, G2, G3, and R1. In one embodiment, G2 and G3 may be trained based on the strength of the connection and information related to communications in the device 10. In one embodiment, through the second training, the importance of canada and travel interests in G2 and G3 may be changed. Referring to the third graph of fig. 4, a similarity related to travel is created in the case of G2, and the importance of interest in canada becomes higher than that of travel in the case of G3. In addition, G1 is linked to B1 via G2, and thus the interest in canada increases in importance.

As described above, the relationship graph between B1, B2, G1, G2, G3, and R1 can be trained by repeated training using information collected from the apparatus 10 and information received from the outside. In one embodiment, when a user writes a message about "travel to canada" by using the device 10, G2 and G3 may be searched for as recommended contacts based on the extracted identifiers "canada" and "travel".

Fig. 5 is a schematic diagram illustrating a method of estimating recommended contacts from a mail of a user by using a user relationship graphic, which is performed by an apparatus, according to an embodiment of the present disclosure.

Referring to fig. 5, in one embodiment, the device 10 may provide a mail service to the user Tom. The device 10 may display mail received from traveling Sunny on day 11/1 of 2017. In one embodiment, the device 10 may identify the identifier by analyzing the content of the mail received from Sunny. For example, the apparatus 10 may analyze text contained in the mail and identify keywords such as "canada", "travel", "Jake", "travel agency", "reply", and the like as identifiers.

In one embodiment, the apparatus 10 may search for recommended contacts based on at least one identifier. In one embodiment, the apparatus 10 may search for recommended contacts by using a user relationship graphic created based on intimacy and matters of interest between users.

In one embodiment, the apparatus 10 may search for Sunny502, Jake 503, Yoon 504, June 505, and travel agency 506 by using the user relationship graph created with Tom 501. For example, Sunny502 may have common interests in canada and may have common interests in traveling with Jake 503.

In one embodiment, the data fields for each user node may include affinity, frequency of contact, location of each user, other interests, and the like. For example, the apparatus 10 may identify location information of each user based on contact with other users, identification information of each user, externally displayed status information, and the like by using the data field.

In one embodiment, the apparatus 10 may identify a recommended contact that is appropriate for the current situation among the searched contacts. In one embodiment, the apparatus 10 may identify the recommended contact based on the current location and the current time. The location and time considered by the apparatus 10 may include a location and time associated with both or each of the user and the contact owner. For example, considering Sunny502 being overseas, device 10 may recommend Yoon 504 in korea instead of Sunny 502.

Alternatively, the apparatus 10 may identify the recommended contact based on the content of the mail. For example, the apparatus 10 may identify a recommendation for Jake 503 based on mail received from Sunny502 traveling with Jake 503, and may identify a recommendation for travel agency 506 from mail received from Sunny502 seeking a travel agency. Otherwise, the device 10 can identify a recommendation for Sunny502 from the email received by Sunny502 that is requesting a reply. Alternatively, the apparatus 10 may identify the recommended contact by considering the mail content and the location information. When it is determined that the user will soon leave canada, the apparatus 10 may recommend Jake 511 of canada residing at the travel destination before recommending Yoon 504 residing in korea.

In one embodiment, the device 10 may display the identified recommended contacts on the screen in the form of a pop-up window 510. In one embodiment, the apparatus 10 may display multiple recommended contacts simultaneously.

In one embodiment, the apparatus 10 may display the attributes of the recommended contacts differently according to a given priority. The device 10 may display the size, color, and degree of deformation of the recommended contact differently according to its priority. In one embodiment, apparatus 10 may display Jake 511 at the first level larger in size than other contacts, or may change the color of Jake 51 so that Jake 511 may be highlighted.

In one embodiment, the device 10 may display Jake 511 at the first level, Sunny512 at the second level, and Yoon 513 at the third level, so that their method of contact links to them. For example, if it is determined that the contact method of Sunny512 at the second level is a mail, the device 10 may execute an application related to the mail when Sunny512 is selected in the pop-up window 510.

Fig. 6 is a schematic diagram illustrating a method of displaying a contact method linked to a selected recommended contact, the method being performed by an apparatus, according to an embodiment of the present disclosure.

Referring to FIG. 6, in one embodiment, when the user selects one of the plurality of recommended contacts, the apparatus 10 may display the contact method of the selected recommended contact.

In one embodiment, device 10 may display Jake's contacts by using pop-up window 610. In one embodiment, the device 10 may display the contact methods according to their priorities.

In one embodiment, the apparatus 10 may obtain the AI training model by using context information of the executing application based on a result of the relationship learning between the executing application and the application installed in the apparatus 10. In one embodiment, the context information of the application may include at least one of environment information of the application, device state information of the application, state information of a user of the application, and application use history information, but is not limited thereto. The environment information of the application is information about an environment within a certain radius range during execution of the application, and may include, but is not limited to, for example, weather information, temperature information, humidity information, illuminance information, noise information, sound information, and the like. The device state information of the application may include, but is not limited to, device mode information (e.g., sound mode, vibration mode, mute mode, power save mode, off mode, multi-window mode, auto-rotate mode, etc.), device location information, device time information, activation information of a communication module (e.g., Wi-Fi on, bluetooth off, Global Positioning System (GPS) on, Near Field Communication (NFC) on, etc.), network connection state information, and the like during execution of the application. The state information of the user is information on the user's movement, life pattern, etc., and may include, but is not limited to, information on the user's walking state, exercise state, driving state, sleeping state, emotional state, etc. The application use history information is information on a user use application history, and may include, but is not limited to, a history of functions performed in the application, a call history of the user, a text message history of the user, and the like.

The device 10 may obtain the AI training model using context information of the application being executed therein based on a result of relationship learning between the application being executed and an application installed therein or an application executed in the device 10 under external control. In one embodiment, the application installed or executed in the apparatus may be an application associated with a contact of another user. For example, the apparatus 10 may be trained using an AI training model such that when a user uses an SNS application outdoors, the user communicates with other users via the same SNS application at a high frequency. Otherwise, the device 10 may be trained using the AI training model such that when the user uses the mail server application at the company, the user communicates with other users via the phone application at a high frequency. However, the above cases are merely examples, and the method of training using the AI training model of the present disclosure is not limited thereto.

In one embodiment, the apparatus 10 may obtain the AI training model based on a relationship between the executing application, the recommended contact, and the application installed in the apparatus 10 by using context information of the executing application and information on communication between a plurality of users.

In one embodiment, the apparatus 10 may identify an application to be linked to at least one recommended contact by using the obtained AI training model. For example, the apparatus 10 may be trained by the AI training model that if the user uses a mail server application at a company, when the recommended contact is a corporate employee, the user communicates with the corporate employee via an in-house messenger application. Otherwise, the device 10 may be trained by the AI training model that if the user uses the mail server application at the company, the user communicates with the client through the phone application when the recommended contact is the client.

In one embodiment, the device 10 may link multiple contact methods to one recommended contact. The plurality of contact methods may include a plurality of applications. In one embodiment, the apparatus 10 may search for a plurality of applications based on the frequency of use, execution time, execution place, execution period, and main function of the application for communicating with the recommended contact.

In one embodiment, the device 10 may apply information regarding communications between multiple users to an AI training model to determine relationships between recommended contacts and applications installed in the device 10. The apparatus 10 may search for a plurality of applications associated with the searched recommended contact by using the searched recommended contact and the learned AI training model.

For example, when the recommended contact is a user with a "business" data field, the device 10 may search an email application as the associated application using the AI training model. Otherwise, when the recommended contact is a user with a "company employee" data field, the apparatus 10 may search for an internal messenger application as the associated application using the AI training model.

In one embodiment, the device 10 may identify the priority of multiple applications.

As shown in fig. 6, in one embodiment, the device 10 may simultaneously display a plurality of contact methods (e.g., an SNS contact method 611, a phone contact method 612, and a mail contact method 613) linked to the selected recommended contact. The device 10 may set attributes of a plurality of contact methods differently according to its priority. For example, when it is recommended that the contact method for communication with Jake is SNS in the present case, the SNS contact method 611 may be displayed in a form of a window larger than the phone contact method 612 or the mail contact method 613, or a color may be displayed in a color different from that of the SNS contact method 611 and the phone contact method 612. Alternatively, the apparatus 10 may display attributes of a plurality of contact methods differently according to their priorities, such as size, color, shape, brightness, or degree of flicker.

In one embodiment, the device 10 may execute a contact method selected by the user from a plurality of contact methods. For example, when the user selects the SNS contact method 61, an SNS application most frequently used by the user and Jake may be executed.

Fig. 7 is a schematic diagram illustrating a method of displaying another contact method linked to a recommended contact when not connected to the recommended contact according to an embodiment of the present disclosure.

Referring to fig. 7, in one embodiment, the apparatus 10 may display at least one recommended contact based on the content of the application currently being executed.

In one embodiment, the device 10 may display the recommended contacts linked to the first level application. The apparatus 10 may display the recommended contacts on a predetermined area 701 of the screen. In one embodiment, the predetermined area 701 may be a lower portion, an upper portion, a middle portion, a left portion, or a right portion of the screen.

In one embodiment, the apparatus 10 may display an icon of an application linked to the recommended contact. For example, the apparatus 10 may display an identification name 703 of the recommended contact, an icon 702 linking the application, and a phone number 704 of the recommended contact in a predetermined area 701 (i.e., a lower portion) of the screen.

In one embodiment, the apparatus 10 may execute an application linked to a recommended contact selected in response to a user input to select the recommended contact. For example, when the user selects the recommended contact, the telephone number 704 corresponding to the identification name 703 of "mother" may be dialed. In this case, the telephone number 704 may be dialed using a function or telephone application provided in the apparatus 10.

In one embodiment, the device 10 may display the second level application when the communication using the first level application for recommending contacts fails. For example, when a phone connection with the mother fails, the mail service may be displayed as a second level application. An icon 705 and mail address 706 of the second level application and an identification name 703 of the recommended contact may be displayed on the screen of the apparatus 10.

Alternatively, the device 10 may execute the second level application upon a communication failure using the first level application. For example, when a phone connection with mother fails, a mail service of sending mail to mother may be performed.

Fig. 8 is a schematic diagram illustrating a method of recommending another recommended contact according to an embodiment of the present disclosure.

Referring to FIG. 8, in one embodiment, the device 10 may display contacts recommended based on the user context on a lower area 801 of its screen. The recommended contact may include Jake as the user's identification name 803 and an icon 802 that links to the application of the recommended contact.

In one embodiment, the apparatus 10 may receive a user input 804 requesting another recommended contact. For example, a user input may be received indicating that the lower region 801 of the screen displaying the recommended contacts is touched. As another example, the user input may include a tap, touch and hold, double click, drag, pan, flick, drag and drop, and the like. User input requesting another recommended contact may be specified by the user.

In one embodiment, in response to a user input 804 requesting another recommended contact, the device 10 may display the other recommended contact on the lower area 801 of its screen. For example, the device 10 may display the identification name 805 as Kim, the mail icon as the icon 807 linked to the application of the recommended contact, and the mail address 806 of Kim.

In one embodiment, the priority of another recommended contact may be lower than the priority of the recommended contact currently being displayed on the screen.

In one embodiment, when displaying the recommending contact, the device 10 may display an advertisement or coupon for an advertised product provided in the recommending contact. In one embodiment, the apparatus 10 may provide convenience in recommending contacts, enhance convenience of life by providing advertisements, and enhance the likelihood of selecting a recommended contact.

Fig. 9 is a schematic diagram illustrating a method of estimating recommended contacts based on schedule information and location information according to an embodiment of the present disclosure.

Referring to FIG. 9, in one embodiment, the appliance 10 may identify recommended contacts using the server 3000 or SNS server 4000 via the network 5000.

The device 10 may provide the server 300 with information about communication established by the device 10 with others, and receive the recommended contact list from the server 3000 and display the list on its screen.

In addition, the device 10 may provide information related to communication, including location information and time information, as well as current location information and schedule information of the user, to the server 300.

The server 300 may create a list of recommended contacts based on schedule information, location information, and time information of the devices 10 belonging to the user. Alternatively, the server 300 may create a list of recommended contacts based on schedule information of the user of the device 10 and acquaintances of the user. In this case, the server 300 may receive schedule information of the user and schedule information of acquaintances of the user from at least one of the device 10 and the SNS server 4000.

The network 5000 may be embodied as a wired network such as a local area network (L AN), a Wide Area Network (WAN) or a Value Added Network (VAN), or a wireless network such as a mobile radio communication network or a satellite communication network the network 5000 is a data communication network having a comprehensive meaning for allowing the network components shown in fig. 9 to smoothly communicate with each other, and the network 5000 includes a wired internet, a wireless internet and a mobile wireless communication network.

Fig. 10 is a diagram illustrating a method of creating an address book after obtaining a new contact, the method being performed by an apparatus, according to an embodiment of the present disclosure.

Referring to fig. 10, in one embodiment, when new contact information is obtained through communication with another user, the apparatus 10 may add the new contact information to the address book based on intimacy, interest, schedule, etc. of the new contact information. In one embodiment, the apparatus 10 may create, modify, and delete address book groups based on relationships with other users.

In one embodiment, device 10 may obtain new contacts I and J. The apparatus 10 may identify trends of the contacts I and J, relationships between the contacts I and J and the user, and public relationships based on the user relationship graph. For example, the apparatus 10 may identify contacts I and J as contacts encountered by the user at a travel meeting. When the address book groups previously stored in the device 10 include a company group 1001, a family group 1002, and a friend group 1003, a travel partner group 1004 may be created to classify contacts I and J. The apparatus 10 may add contacts I and J to the newly created travel partner group 1004.

Fig. 11 is a schematic diagram illustrating a method of updating an address book based on a user relationship graph, the method being performed by an apparatus, according to an embodiment of the present disclosure.

Referring to FIG. 11, in one embodiment, the apparatus 10 may update the user relationship graphic based on information regarding communications between the user and other users. In one embodiment, the apparatus 10 may sort the address book based on the updated user relationship graph.

In one embodiment, device 10 may identify that a personal relationship is established with contact B1101 classified as belonging to corporate group 1001 based on information about communications between the user and contact B1101. The device 10 may recognize that contact B1101 is similar to the user included in the group of friends 1003. Device 10 may move contact B1101 from the corporate group 1001 to the friends group 1003. The user may identify contact B1102 belonging to the group of friends 1003 in the address book of the device 10.

Thus, in one embodiment, the address book may be automatically arranged based on information regarding communications between the user and other users.

Fig. 12 is a schematic diagram illustrating recommendation of a message based on schedule information and group information, which is performed by an apparatus according to an embodiment of the present disclosure.

Referring to FIG. 12, in one embodiment, the device 10 may obtain calendar information for the user and identify an application linked to the recommended contact based on the obtained calendar information. For example, the apparatus 10 may obtain information "birthday of 2018.09.04H" from the schedule information of the user. Device 10 may recommend H's contact on 9/4/2018 as shown at 1201. In this case, the device 10 may display that an SNS message 1203 is to be sent to H, reflecting the fact that SNS is a contact method commonly used for communicating with H.

In one embodiment, the apparatus 10 may provide a document form in which a tone, syntax, and sentence structure are predetermined based on the affinity determined for the recommended contacts. For example, the device 10 may provide the phrase "i am loved" 1204 in the SNS message 1203 based on the fact that the user often calls H "i am loved" in a conversation with H, and provide the phrase "birthday happy" based on the fact that 9 months and 4 days are the birthday of H.

In one embodiment, the apparatus 10 may provide a predetermined document form based on the group attribute of the address book. For example, the device 10 may obtain schedule information "2018.09.24 Chuseok (mid-autumn). In one embodiment, device 10 may transmit messages related to each group's mid-autumn festival greeting on 24/9/2018, as shown at 1202. In one embodiment, the apparatus 10 may send a public tone message to the corporate group using a text message 1205. In another embodiment, the device 10 may send a friend message to a group of friends using an SNS message 1206 instead of the text message 1205 sent to the corporate group.

In one embodiment, the apparatus 10 may automatically send a recommendation message to the recommending contact based on the schedule information.

Fig. 13 is a block diagram of an apparatus according to an embodiment of the present disclosure.

Fig. 14 is a detailed block diagram of an apparatus according to an embodiment of the present disclosure.

Referring to fig. 13, in one embodiment, the apparatus 10 may include a memory 1100, a display 1210, and a processor 1300.

However, all of the components shown in FIG. 13 are not essential components of the device 10. The apparatus 10 may also include other components, as well as those shown in fig. 13, or may include only some of those shown in fig. 13.

Referring to fig. 14, the apparatus 10 according to an embodiment may further include an output interface 1200, a communication unit 1500, a sensing unit 1400, an audio/video (a/V) input interface 1600, and a user input interface 1700, as well as a memory 1100, a display 1210, a camera 1610, and a processor 1300.

The memory 1100 may store programs for processing and control executed by the processor 1300, and may store images input to the apparatus 10 or guide information output from the apparatus 10. In addition, the memory 1100 may store information for determining whether to output the booting information.

The memory 1100 may include at least one type of storage medium among a flash memory type storage medium, a hard disk type storage medium, a multimedia card micro type storage medium, a card type memory (e.g., SD or XD memory, etc.), a Random Access Memory (RAM), a static RAM (sram), a Read Only Memory (ROM), an Electrically Erasable Programmable Read Only Memory (EEPROM), a programmable ROM (prom), a magnetic memory, a magnetic disk, and an optical disk.

The program stored in the memory 1100 may be classified into a plurality of modules according to its function, for example, a User Interface (UI) module 1110, a touch screen module 1120, a notification module 1130, and the like.

The UI module 1110 may provide a dedicated UI, a Graphical User Interface (GUI), and the like linked to the apparatus 10 in units of applications. The touch screen module 1120 may sense a touch gesture on the touch screen of the user and transmit information about the touch gesture to the processor 1300. In one embodiment, the touch screen module 1120 may recognize and analyze touch codes. The touch screen module 1120 may be configured as separate hardware including a controller.

The notification module 1130 may generate a signal for notifying the occurrence of an event occurring in the apparatus 10. Examples of events occurring in the device 10 include call signal reception, message reception, key signal input, schedule notification, and the like. The notification module 1130 may output a notification signal in the form of a video signal through the display 1210, an audio signal through the audio output interface 1220, or a vibration signal through the vibration motor 1230. For example, the notification module 1130 may generate a signal for outputting guidance information based on the estimated lane information.

The output interface 1200 may output an audio signal, a video signal, or a vibration signal. The output interface 1200 may include a display 1210, an audio output interface 1220, and a vibration motor 1230.

The display 1210 displays and outputs information processed by the apparatus 10. In detail, the display 1210 may output an image captured by the camera 1610. Further, the display 1210 may combine the guide information created by the processor 1300 with the captured image and output the combined result.

In addition, the display 1210 may display a user interface for performing an operation in response to a user input.

The audio output interface 1220 outputs audio data received from the communication unit 1500 or stored in the memory 1100. Further, the audio output interface 1220 outputs a sound signal (for example, a call signal reception sound, a message reception sound, or a notification sound) related to a function performed by the apparatus 10. For example, the audio output interface 1220 may output the guide information, which is generated in the form of a signal by the notification module 1130, in the form of a sound signal under the control of the processor 1300.

In general, the processor 1300 controls the overall operation of the device 10. For example, the processor 1300 may execute programs stored in the memory 1100 to comprehensively control the user input interface 1700, the output interface 1200, the sensing unit 1400, the communication unit 1500, the a/V input interface 1600, and the like. In addition, the processor 1300 may perform the functions of the apparatus 10 by executing the program stored in the memory 1100.

The sensing unit 1400 may sense a state of the apparatus 10 or a state of the surrounding environment of the apparatus 10 and transmit information about the sensed state to the processor 1300.

The sensing unit 1400 may include, but is not limited to, at least one of a geomagnetic sensor 1410, an acceleration sensor 1420, a temperature/humidity sensor 1430, an infrared sensor 1440, a gyro sensor 1450, a position sensor (e.g., a Global Positioning System (GPS))1460, a barometer sensor 1470, a proximity sensor 1480, and a red-green-blue (RGB) sensor 1490. The function of the sensors can be intuitively inferred by those skilled in the art from the names of these sensors and will not be described in detail herein.

In one embodiment, the sensing unit 1400 may measure a distance between at least one object identified from the captured image and the vehicle.

The communication unit 1500 may include one or more components configured to enable the apparatus 10 to communicate with another apparatus (not shown) and a server (not shown). Other devices may be, but are not limited to, computing devices or sensing devices similar to device 10. For example, the communication unit 1500 may include a short-range wireless communication unit 1510, a mobile communication unit 1520, and a broadcast receiving unit 1530.

Examples of the short-range wireless communication unit 1510 may include, but are not limited to, a bluetooth communication unit, a bluetooth low energy (B L E) communication unit, a near field communication unit, a W L AN (Wi-Fi) communication unit, a ZigBee communication unit, AN infrared data association (IrDA) communication unit, a Wi-Fi direct (WFD) communication unit, AN Ultra Wideband (UWB) communication unit, AN adaptive network technology (Ant +) communication unit, and the like.

The mobile communication unit 1520 transmits or receives a radio signal to or from at least one of a base station, an external terminal, and a server in a mobile communication network. Here, the radio signal may be understood to include a voice call signal, a video call signal, or various types of data generated when transmitting and receiving text/multimedia messages.

The broadcast receiving unit 1530 receives a broadcast signal and/or broadcast associated information from the outside via a broadcast channel. The broadcast channel may include a satellite channel, a terrestrial channel. In one embodiment, the apparatus 10 may not include the broadcast receiving unit 1530.

The a/V input interface 1600 is configured to input an audio signal or a video signal, and may include a camera 1610, a microphone 1620, and the like. The camera 1610 may obtain video frames, such as still images or moving images, through an image sensor in a video call mode or a photographing mode. Images captured via the image sensor may be processed by the processor 1300 or another image processor (not shown).

In one embodiment, the camera 1610 may capture images of the exterior of the vehicle. For example, the camera 1610 may capture an image of the surroundings in front of the vehicle while driving, but the embodiment is not limited thereto.

The microphone 1620 receives an external audio signal and converts the audio signal into electronic voice data. For example, the microphone 1620 may receive an audio signal from an external device or a user. The microphone 1620 may remove noise generated when receiving an external audio signal using various noise suppression algorithms.

User input interface 1700 may be understood to represent a device through which a user inputs data for controlling device 10. Examples of user input interface 1700 may include, but are not limited to, a keyboard, dome switch (dome switch), touch panel (touch capacitive touch panel, pressure resistive overlay touch panel, infrared sensor touch panel, surface acoustic wave conductive touch panel, integrated strain gauge touch panel, piezoelectric effect touch panel, etc.), jog dial, rotary switch, and the like.

Referring to FIG. 15, in one embodiment, processor 1300 may include a data learner 1310 and a data identifier 1320.

The data learner 1310 may learn a graph showing relationships between the user and other users based on information related to communications between the user and other users. The data learner 1310 may learn the type of data to be used, thereby increasing the accuracy of the relationship graph and whether to extend the relationship graph with the data. The data learner 1310 may obtain data to be used for learning and learn criteria for forming an appropriate relationship network by learning a relationship graph that applies the obtained data to a data recognition model (described below).

In one embodiment, when there are a plurality of pre-established data recognition models, the data learner 1310 may identify a data recognition model that is highly correlated with the input training data and the base training data as the data recognition model to be learned. In this case, the basic training data may be classified in advance according to the data type, and a data recognition model may be established in advance for each type of data. For example, the basic training data may be classified in advance by various criteria, such as a place where the training data is created, a time when the training data is created, a size of the training data, a kind of the training data, a creator of the training data, a type of an object included in the training data, and the like.

The data identifier 1320 may increase the accuracy of contact recommendations based on the relationship graph of the user, as the case may be. Based on the results of learning the relationship between the context identifier and the pieces of contact information by using the learned data recognition model, the data identifier 1320 may increase the accuracy of the contact recommendation. The data identifier 1320 may obtain data according to a predetermined standard by learning and use a data recognition model using the obtained data as an input value to improve the accuracy of the contact recommendation. The resulting value output of the data recognition model using the obtained data as input values may be used to update the data recognition model.

At least one of the data learner 1310 and the data identifier 1320 may be fabricated in the form of at least one hardware chip and installed in the apparatus. For example, at least one of the data learner 1310 and the data identifier 1320 may be manufactured in the form of a dedicated hardware chip for Artificial Intelligence (AI), or may be manufactured as a part of an existing general-purpose processor (e.g., a Central Processing Unit (CPU) or an Application Processor (AP)) or a part of a dedicated graphics processor (e.g., GPU) and then installed in various types of devices as described above.

In this case, the data learner 1310 and the data identifier 1320 may be installed in one device or different devices. For example, one of the data learner 1310 and the data identifier 1320 may be included in the apparatus, and the other may be included in the server. The data learner 1310 and the data identifier 1320 may be connected to each other by wire or wirelessly, so that information about the model established by the data learner 1310 may be provided to the data identifier 1320, and data input to the data identifier 1320 may be provided to the data learner 1310 as additional training data. At least one of the data learner 1310 and the data identifier 1320 may be embodied as a software module. When one of the data learner 1310 and the data identifier 1320 may be embodied as a software module (or a program module including instructions), the software module may be stored in a non-transitory computer readable recording medium. In this case, the at least one software module may be provided by an Operating System (OS) or by some application. Alternatively, a part of at least one software module may be provided by the OS and another part may be provided by an application.

The data identifier 1320 may be trained based on the training data using a data recognition model for determining a situation. In this case, the data recognition model may be a pre-constructed model. For example, the data recognition model may be a model that is previously constructed by receiving basic training data (e.g., sample images, etc.). The data recognition model may be constructed in advance by considering an application field of the recognition model, a learning purpose or a computer performance of the device, and the like. The data recognition model may be, for example, a neural network-based model. For example, a model such as a Deep Neural Network (DNN), a Recurrent Neural Network (RNN), or a Bidirectional Recurrent Deep Neural Network (BRDNN) may be used as the data recognition model, but the embodiment is not limited thereto.

In various embodiments, when there are multiple pre-constructed data recognition models, the data learner 1310 may identify a data recognition model that is highly correlated with the received training data and the base training data as the data recognition model to be learned. In this case, the basic training data may be classified in advance according to the data type, and the data recognition model may be constructed in advance according to the data type. For example, the basic training data may be classified in advance according to various criteria such as a place where the training data is created, a time when the training data is created, a size of the training data, a kind of the training data, a creator of the training data, and a type of an object included in the training data.

Embodiments may be implemented in the form of computer-readable recording media, such as computer-executable program modules, that store instructions that are executable by a computer. The computer-readable recording medium may be any available medium that can be accessed by the computer, and examples thereof include a volatile recording medium, a nonvolatile recording medium, a detachable storage medium, and a non-detachable recording medium. Examples of the computer readable recording medium may also include computer storage media and communication media. Examples of the computer readable recording medium may include volatile recording media, nonvolatile recording media, removable recording media, and non-removable recording media manufactured by any method or technology for storage of information such as computer readable instructions, data structures, program modules, or other data. In general, examples of communication media include computer readable instructions, data structures, program modules, or other data in a modified data signal, other transmission mechanisms, or any information transmission media.

In the present disclosure, the term "unit" may be understood to mean a hardware component, such as a processor or a circuit, and/or a software component executable by a hardware component, such as a processor.

The embodiments set forth herein are intended to provide examples, and it will be apparent to those of ordinary skill in the art that various changes may be easily made in the embodiments without departing from the technical idea and essential features of the present disclosure. Accordingly, the embodiments set forth herein should be considered in descriptive sense only and not for purposes of limitation. For example, components described herein as being separate from one another may be implemented in combination with one another, and similarly, components described herein as being combined with one another may be implemented as being separate from one another.

While the disclosure has been shown and described with reference to various embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the disclosure as defined by the appended claims and their equivalents.

Claims

1. An apparatus, comprising:

a processor; and

a memory configured to store instructions executable by the processor, wherein the processor is configured to execute the instructions to:

extracting context information from the displayed data based on an application being executed by the apparatus,

an identifier is identified from the context information,

searching for at least one recommended contact associated with the identifier,

identifying a priority of the at least one recommended contact, an

And controlling to display the at least one recommended contact according to the priority.

2. The apparatus of claim 1, wherein the processor is further configured to:

searching for the at least one recommended contact related to the identifier based on the identifier and a relationship graph obtained by inputting information about communications between a plurality of users to a first training model for determining associations between the plurality of users,

wherein the association is represented by a predicate having a pair of data or nodes as subject and object, respectively.

3. The apparatus of claim 1, wherein the processor is further configured to:

controlling to display a first level of recommended contacts according to the priority of the at least one recommended contact, an

And when the communication with the recommended contact person of the first level fails, controlling to display the recommended contact person of the second level.

4. The apparatus of claim 2, wherein the relationship graph comprises:

a plurality of nodes corresponding to the plurality of users; and

at least one line representing a relationship between the plurality of users, wherein a thickness of the at least one line represents a strength of the relationship between the plurality of users.

5. The apparatus of claim 4, wherein the first training model is configured to:

analyzing information on a usage history of the identifier among the information on the communication,

assigning a weight to the used contact information according to an application using the identifier, a use period, a time using the identifier, a number of times using the identifier, and a place of use, and

determining the at least one recommended contact based on the assigned weight.

6. The apparatus of claim 4, wherein the processor is further configured to create an address book by identifying affinities among a plurality of pieces of contact information using the first training model and classifying the plurality of pieces of contact information into a plurality of groups based on the identified affinities.

7. The apparatus of claim 4, wherein when obtaining new contact information, the processor is further configured to add an address book by classifying the new contact information into one of a plurality of groups based on information regarding communications with a user having the new contact information.

8. The apparatus of claim 7, wherein the processor is further configured to reclassify a set of contact information of a plurality of pieces of contact information for which affinity has changed.

9. The apparatus of claim 6, wherein the processor is further configured to:

obtaining user schedule information, an

Controlling to display all contact information belonging to one of the plurality of groups as recommended contact information based on the obtained user schedule information.

10. The apparatus of claim 2, wherein the processor is further configured to identify which application to link to the at least one recommended contact by applying information about the communication between the plurality of users to a second training model for determining a relationship between the at least one recommended contact and an application in the apparatus.

11. The apparatus of claim 10, wherein the processor is further configured to execute an application linked to a recommended contact selected in response to a user input selecting one of the at least one recommended contact.

12. The apparatus of claim 10, wherein the processor is further configured to:

identifying a plurality of applications to be linked to one recommended contact, identifying priorities of the plurality of applications based on a frequency of use, an execution time, an execution place, an execution period, and a main function of an application for communicating with the one recommended contact, and

when the user selects the recommended contact, the plurality of applications are controlled to be displayed according to the priorities of the plurality of applications.

13. The apparatus of claim 12, wherein the processor is further configured to:

executing a first level of application when the user selects the recommended contact, an

And executing the application of the second level when the communication between the application of the first level and the recommended contact person fails.

14. The apparatus of claim 10, wherein the processor is further configured to:

obtaining user schedule information, an

Identifying an application to be linked to the at least one recommended contact based on the obtained user schedule information.

15. A method, comprising:

extracting context information from the displayed data based on an application being executed by the apparatus;

identifying an identifier from the context information;

searching for at least one recommended contact associated with the identifier;

determining the priority of the searched at least one recommended contact; and

and displaying the at least one recommended contact according to the priority.