Method, device and equipment for recommending missing person based on address book and storage medium

Abstract

The present application relates to artificial intelligence technology, disclose a kind of based on address book's person searching recommendation method, device, equipment and storage medium.The method comprises: obtaining the voice data of user and text operation is carried out, obtain text sentence, extract the keyword in the text sentence, and utilize keyword set to query pre-constructed address book database, obtain person searching result set, obtain the knowledge graph of user and the recommendation knowledge graph of each person searching result in person searching result set, text clustering is carried out to the user knowledge graph and each the recommendation knowledge graph, obtain the clustering score corresponding to the user and each recommended character of the user;According to the clustering score and the preset recommendation sorting strategy, the sorting operation based on name character length and clustering relationship is carried out to each recommended character, and recommendation sorting result is obtained.The present application can increase the accuracy of address book person searching.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence technology, and in particular to a method, apparatus, device, and computer-readable storage medium for finding and recommending people based on an address book. Background Technology

[0002] With the development of society, there are more and more large enterprises, and the increase in the number of employees in these enterprises has led more and more companies to use intelligent contact search methods for personnel lookup. However, contact search targets simply extract names for display. When encountering people with the same name from different organizations, the names found are often limited and fixed, requiring users to carefully select, and cannot quickly find the information of the user's desired target person. Summary of the Invention

[0003] This invention provides a method, apparatus, device, and storage medium for finding people based on a contact list, the main purpose of which is to increase the accuracy of finding people using a contact list.

[0004] To achieve the above objectives, the present invention provides a method for finding and recommending people based on a contact list, comprising:

[0005] Acquire user-input voice data and convert the voice data to text to obtain text statements;

[0006] Keyword extraction is performed on the text statement to obtain a keyword set, and the pre-built address book database is queried based on the keyword set to obtain a set of missing person results;

[0007] Obtain user information and use the user information to construct a user knowledge graph; obtain target information of each recommended person in the missing person result set and construct a recommendation knowledge graph corresponding to each recommended person based on the target information.

[0008] The user knowledge graph and each of the recommendation knowledge graphs are subjected to text clustering to obtain the clustering scores corresponding to the user and each of the recommended individuals.

[0009] Based on the preset recommendation ranking strategy and the clustering score, the recommended individuals are ranked according to their name character length and clustering relationship to obtain the recommendation ranking result.

[0010] Optionally, constructing a user knowledge graph using the user information includes:

[0011] The feature extraction network of the pre-built graph construction model is used to perform entity extraction, relationship recognition and attribute recognition operations on the user information to obtain entity set, relationship set and attribute set;

[0012] The graph construction network of the model is constructed using the graph, and entity linking operations are performed according to the correspondence between the entity set, the relation set, and the attribute set to obtain a knowledge representation relation set;

[0013] According to a preset knowledge fusion strategy, ambiguity is eliminated for each knowledge expression relation in the knowledge expression relation set to obtain a fact expression relation set.

[0014] Based on the preset conceptual framework, hierarchical identification and entity classification operations are performed on each entity in the fact expression relationship set. Based on the hierarchical identification results and entity classification results, the fact expression relationship set is arranged in sequence to obtain the user knowledge graph.

[0015] Optionally, the step of performing text clustering on the user knowledge graph and each of the recommendation knowledge graphs to obtain clustering scores corresponding to the user and each of the recommendation figures includes:

[0016] Using the cosine similarity algorithm, the similarity between the user knowledge graph and each recommendation knowledge graph is calculated based on multiple entity types to obtain the similarity scores of multiple entity types between the user and each recommendation person.

[0017] The knowledge graph is weighted according to preset weight coefficients, and the similarity scores of the multiple entity types are weighted according to the weight configuration relationship to obtain the clustering scores of the user and each recommended person.

[0018] Optionally, the step of sorting each recommended person based on name character length and clustering relationship according to a preset recommendation ranking strategy and the clustering score to obtain the recommendation ranking result includes:

[0019] Obtain the name length of each recommended person to get the name character length;

[0020] According to a preset recommendation arrangement strategy, the name character length is grouped to obtain multiple name recommendation groups, and the name recommendation groups are arranged in order from longest to shortest name character length.

[0021] Based on the cluster scores of each recommended person from highest to lowest, the recommended groups of names are arranged within each group to obtain the recommendation ranking result.

[0022] Optionally, the step of extracting keywords from the text statement to obtain a keyword set includes:

[0023] The text statement is segmented using a word segmentation tool to obtain a set of segmentation results.

[0024] The word segmentation result set is cleaned by deleting stop words from the preset stop word dictionary.

[0025] The word sense category of each word in the clean word segmentation result set is identified by using a pre-trained target entity recognition model to obtain the category recognition result, and word segments of a preset type are extracted from the category recognition result to obtain a keyword set.

[0026] Optionally, the step of querying a pre-built address book database based on the keyword set to obtain a set of search results includes:

[0027] Based on the keyword set, the user's historical communication records are precisely matched to obtain a priority result set;

[0028] A fuzzy search is performed on the target type strings in the keyword set to the address book database, and an exact match is performed on the keyword characters other than the target type strings in the keyword set to the address book database to obtain a preliminary set of missing person results.

[0029] Based on the priority result set, a portion of the search results in the primary search result set are prioritized to obtain the search result set.

[0030] Optionally, after performing a sorting operation on each recommended person based on name character length and clustering relationship to obtain the recommendation ranking result, the method further includes:

[0031] When a user selects a target person from the recommended ranking results, the target person and the user are stored in a preset query response count table;

[0032] By using a pre-defined data extraction whitelist, the social relationships of target users across multiple target communication applications can be obtained.

[0033] The user knowledge graph is dynamically updated based on the social relationships and the query response count table.

[0034] To address the above problems, the present invention also provides a contact-based person-finding recommendation device, the device comprising:

[0035] The text acquisition module is used to acquire the voice data input by the user and convert the voice data into text to obtain text sentences;

[0036] The keyword-based person search module is used to extract keywords from the text statement to obtain a keyword set, and then query a pre-built address book database based on the keyword set to obtain a set of person search results.

[0037] The knowledge graph construction module is used to obtain user information and construct a user knowledge graph using the user information, and to obtain target information of each recommended person in the search results set and construct a recommendation knowledge graph corresponding to each recommended person based on the target information.

[0038] The clustering query module is used to perform text clustering on the user knowledge graph and each of the recommendation knowledge graphs to obtain the clustering scores corresponding to the user and each of the recommended individuals, and to perform a sorting operation on each of the recommended individuals based on the name character length and clustering relationship according to the preset recommendation sorting strategy and the clustering scores to obtain the recommendation sorting results.

[0039] To address the above problems, the present invention also provides an electronic device, the electronic device comprising:

[0040] At least one processor; and,

[0041] A memory communicatively connected to the at least one processor; wherein,

[0042] The memory stores a computer program that can be executed by the at least one processor, which enables the at least one processor to perform the address book-based person-finding recommendation method described above.

[0043] To address the aforementioned problems, the present invention also provides a computer-readable storage medium storing at least one computer program, which is executed by a processor in an electronic device to implement the aforementioned address book-based person-finding recommendation method.

[0044] This invention provides a method for finding people in a contact list by converting their text to text and extracting keywords to identify a keyword set. This enables basic contact search operations and yields a result set. Then, by identifying the user's and recommended individuals' information, a user knowledge graph and various recommended knowledge graphs are generated. These knowledge graphs are then clustered to obtain clustering scores, which represent the proximity of the relationship between each recommended individual and the user, allowing for prioritized recommendations. Furthermore, the result set incorporates fuzzy search to avoid keyword errors caused by speech-to-text conversion. Therefore, the result set may contain multiple names. Finally, the recommended individuals are sorted based on name character length and clustering relationships to obtain a ranking result, enabling personalized search results and more accurate results. Therefore, this invention provides a contact list-based contact search recommendation method, apparatus, device, and storage medium that improves the accuracy of contact list-based contact search. Attached Figure Description

[0045] Figure 1 This is a flowchart illustrating a contact-based person-finding and recommendation method according to an embodiment of the present invention.

[0046] Figure 2 This is a detailed flowchart illustrating one step of a contact-based person-finding and recommendation method according to an embodiment of the present invention.

[0047] Figure 3 This is a detailed flowchart illustrating one step of a contact-based person-finding and recommendation method according to an embodiment of the present invention.

[0048] Figure 4 This is a detailed flowchart illustrating one step of a contact-based person-finding and recommendation method according to an embodiment of the present invention.

[0049] Figure 5 This is a functional block diagram of a contact list-based person-finding and recommendation device provided in an embodiment of the present invention;

[0050] Figure 6 This is a schematic diagram of the structure of an electronic device that implements the address book-based person-finding and recommendation method according to an embodiment of the present invention.

[0051] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0052] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0053] This application provides a method for finding and recommending missing persons based on a contact list. In this application, the executing entity of the method includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, etc. In other words, the method for finding and recommending missing persons based on a contact list can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0054] Reference Figure 1 The diagram shown is a flowchart illustrating a contact list-based person-finding recommendation method according to an embodiment of the present invention. In this embodiment, the contact list-based person-finding recommendation method includes:

[0055] S1. Obtain the voice data input by the user, and convert the voice data to text to obtain text statements.

[0056] In this embodiment of the invention, when a user performs a search operation in the intelligent FAQ address book, the user's voice data is automatically acquired, and then the voice data is converted into text statements through a voice-to-text operation.

[0057] S2. Extract keywords from the text statement to obtain a keyword set, and query a pre-built address book database based on the keyword set to obtain a set of missing person results.

[0058] In this embodiment of the invention, the keywords include information such as personal names, place names, departments, and attributes.

[0059] In detail, in this embodiment of the invention, the step of extracting keywords from the text statement to obtain a keyword set includes:

[0060] The text statement is segmented using a word segmentation tool to obtain a set of segmentation results.

[0061] The word segmentation result set is cleaned by deleting stop words from the preset stop word dictionary.

[0062] The word sense category of each word in the clean word segmentation result set is identified by using a pre-trained target entity recognition model to obtain the category recognition result, and word segments of a preset type are extracted from the category recognition result to obtain a keyword set.

[0063] In this embodiment of the invention, the text statement can be segmented using word segmentation tools such as jieba or Chinese word segmentation systems to obtain a set of segmentation results.

[0064] Furthermore, the stop words refer to words that appear repeatedly but have no practical meaning and easily consume recognition computing resources, such as "ah" and "this." In this embodiment of the invention, a pre-constructed stop word dictionary can be used to delete each stop word from the segmentation result set, resulting in a clean segmentation result set. Then, a pre-trained target entity recognition model is used to perform part-of-speech tagging on the clean segmentation result set, and then each adjective and noun is identified as a keyword to obtain a keyword set. The target entity recognition model in this embodiment is a neural network based on a combination of BERT and CRF; its training process is not described in detail.

[0065] It should be understood that during the speech-to-text process, compilation errors may occur due to the user's accent. Therefore, the text statements are not 100% accurate, and words such as place names are less prone to translation errors due to their frequent use, while custom-defined personal names are more prone to errors. Therefore, to expand the scope of finding people in the address book, this embodiment of the invention uses a fuzzy matching method to perform fuzzy searches on various keywords to obtain a set of search results.

[0066] In detail, in this embodiment of the invention, the step of querying a pre-built address book database based on the keyword set to obtain a set of search results includes:

[0067] Based on the keyword set, the user's historical communication records are precisely matched to obtain a priority result set;

[0068] A fuzzy search is performed on the target type strings in the keyword set to the address book database, and an exact match is performed on the keyword characters other than the target type strings in the keyword set to the address book database to obtain a preliminary set of missing person results.

[0069] Based on the priority result set, a portion of the search results in the primary search result set are prioritized to obtain the search result set.

[0070] In this embodiment of the invention, after obtaining the keyword set, a precise match is performed on historical communication records to check if a call has been made. If a call has been made, the recommendation priority is increased. Here, precise matching refers to matching the pinyin strings of names in the keyword set character by character without error, and also matching information such as departments and companies in the keyword set. In this embodiment, the pinyin results of custom-type keywords such as group names and personal names are used as target type strings to obtain the remaining keyword characters in the keyword set other than the target type strings, such as "Guangzhou Shenzhen, **bank," etc.

[0071] In this embodiment of the invention, querying historical communication records requires exact matching, while querying the address book database can expand the query scope using fuzzy search. Specifically, when querying the address book database, only the target type string is subjected to fuzzy search, while other strings are still matched precisely. Fuzzy search means that any string containing the keyword character can be selected. For example, when the keyword character is "jinchen", strings such as "jincheng" and "wangjinchen" can be selected, thereby increasing the query scope.

[0072] This invention, through querying a contact database, can obtain a combination of search results. Furthermore, querying historical communication records can increase the priority of the query results. Therefore, based on the priority result set, a portion of the search results in the initial search result set is prioritized to obtain the final search result set. The priority marking can be implemented using special symbols or by changing font color, etc.

[0073] S3. Obtain user information and use the user information to construct a user knowledge graph, and obtain target information of each recommended person in the search result set, and construct a recommendation knowledge graph corresponding to each recommended person based on the target information.

[0074] In this embodiment of the invention, the user information and the target information of each recommended person can be retrieved from the enterprise database. The user information and target information refer to the information set other than the keyword set, which can reduce the order of magnitude of the knowledge graph and improve query efficiency.

[0075] Then, by constructing a knowledge graph of users and each recommender, it is easier to organize information about users and each recommender and to find out the relationship between each recommender and the user.

[0076] For details, please refer to the following: Figure 2As shown in this embodiment of the invention, constructing a user knowledge graph using the user information includes:

[0077] S31. Using the feature extraction network of the pre-built graph construction model, entity extraction, relation recognition and attribute recognition operations are performed on the user information to obtain entity set, relation set and attribute set;

[0078] S32. Using the graph to construct the model's graph construction network, perform entity linking operations based on the correspondence between the entity set, the relation set, and the attribute set to obtain a knowledge representation relation set;

[0079] S33. According to the preset knowledge fusion strategy, the ambiguity of each knowledge expression relation in the knowledge expression relation set is eliminated to obtain the fact expression relation set;

[0080] S34. Based on the preset conceptual framework, perform hierarchical identification and entity classification operations on each entity in the fact expression relationship set, and arrange the fact expression relationship set in sequence according to the hierarchical identification results and entity classification results to obtain the user knowledge graph.

[0081] The graph construction model is a neural network model based on a feature extraction network and a graph construction network. The graph construction network is a network that does not require training; therefore, it can be used simply by obtaining the trained feature extraction network.

[0082] In this embodiment of the invention, the conceptual framework refers to constructing a knowledge graph using an entity-relationship-entity or entity-relationship-attribute structure. Entities can be obtained through named entity recognition (NER), relationships can be identified through pattern matching, and attributes are obtained by mining the relationships between entity attributes and their values ​​using data mining methods. Then, according to the conceptual framework, the entity set, the relationship set, and the attribute set are combined to obtain knowledge representation relationships of entity-relationship-entity or entity-relationship-attribute structures, resulting in a knowledge representation relationship set.

[0083] Furthermore, the knowledge fusion strategy refers to methods for logically attributing and filtering redundancy / errors in the data. Because some entities in the knowledge representation relationship set may have the same name but be different things, or the same thing but different names, it is necessary to clarify each knowledge representation relationship through clustering or semantic judgment to obtain a factual representation relationship set.

[0084] Finally, the network constructs the graph to obtain the categories of each entity, such as "person," "thing," and "event," and the hierarchical relationships of each entity, based on the set of factual expression relationships. This allows for the sequential arrangement of the set of factual expression relationships, resulting in the user knowledge graph. The construction process for the recommendation knowledge graphs corresponding to each recommended person is similar and will not be elaborated upon here.

[0085] S4. Perform text clustering on the user knowledge graph and each of the recommendation knowledge graphs to obtain the clustering scores corresponding to the user and each of the recommendation figures.

[0086] In this embodiment of the invention, clustering can be used to identify the clustering scores of each recommender and the user, thereby confirming the closeness of their relationship.

[0087] For details, please refer to the following: Figure 3 As shown in this embodiment of the invention, the step of performing text clustering on the user knowledge graph and each of the recommendation knowledge graphs to obtain the clustering scores corresponding to the user and each of the recommendation figures includes:

[0088] S41. Using the cosine similarity algorithm, perform similarity calculations based on multiple entity types on the user knowledge graph and each recommendation knowledge graph to obtain the similarity scores of multiple entity types corresponding to the user and each recommendation person.

[0089] S42. Configure weights for different entity types in the knowledge graph according to preset weight coefficients, and calculate the similarity scores of the multiple entity types according to the weight configuration relationship to obtain the clustering scores corresponding to the user and each recommended person.

[0090] The cosine similarity algorithm uses the cosine of the angle between two vectors in a vector space as a measure of the difference between two individuals. In this embodiment, word vector encoding techniques, such as the BERT model, can quantify each entity, relation, and attribute in the knowledge graph to obtain a unique quantified code. Then, the cosine similarity algorithm is used to calculate the similarity score between each recommended person and the user. However, different entity types have varying degrees of influence on the calculation of the closeness of relationships. For example, the influence of being in the same department is greater than that of being from the same place of origin, which is greater than that of being the same in height, gender, and age. Therefore, this embodiment uses preset weight coefficients to configure weights for different entity types in the knowledge graph, and calculates the weighted similarity scores of multiple entity types according to the weight configuration relationship. This yields a weighted sum of the similarity scores between the user and each recommended person, and the weighted sum is used as the clustering score.

[0091] S5. Based on the preset recommendation ranking strategy and the clustering score, perform a ranking operation on each recommended person according to the name character length and clustering relationship to obtain the recommendation ranking result.

[0092] In this embodiment of the invention, the recommendation ranking strategy is to first sort the recommended individuals by name length, and then sort them by clustering score.

[0093] For details, please refer to the following: Figure 4 As shown in this embodiment of the invention, the step of sorting each recommended person based on name character length and clustering relationship according to a preset recommendation ranking strategy and the clustering score to obtain the recommendation ranking result includes:

[0094] S51. Obtain the name length of each recommended person to get the name character length;

[0095] S52. According to the preset recommendation arrangement strategy, the name character length is grouped to obtain multiple name recommendation groups, and the name recommendation groups are arranged in order from longest to shortest name character length.

[0096] S53. Based on the clustering scores of each recommended person from highest to lowest, the recommended name groups are arranged within each group to obtain the recommendation ranking result.

[0097] In this embodiment of the invention, firstly, long names are prioritized and placed at the top. Secondly, based on this, the recommendations are sorted according to the user's own characteristics and the clustering scores of each recommended person to obtain the recommendation ranking result. For example, if a user enters "I want to find A", and the user belongs to Bank B, then the system will prioritize recommending A from Bank B.

[0098] Furthermore, in another embodiment of the present invention, after performing a sorting operation on each recommended person based on name character length and clustering relationship to obtain the recommendation ranking result, the method further includes:

[0099] When a user selects a target person from the recommended ranking results, the target person and the user are stored in a preset query response count table;

[0100] By using a pre-defined data extraction whitelist, the social relationships of target users across multiple target communication applications can be obtained.

[0101] The user knowledge graph is dynamically updated based on the social relationships and the query response count table.

[0102] The number of responses in the query response count table indicates a user's historical communication history, while the social relationships indicate a user's recent communication history. A whitelist can be constructed by extracting data from a preset database, identifying individuals with call records within the recent time period for each target communication application. Both the social relationships and the query response count table play a crucial role in improving the accuracy of missing person recommendations. In this embodiment of the invention, the user knowledge graph can be dynamically updated using the number of query responses in the query response count table and the social relationships, thereby increasing the accuracy of missing person searches.

[0103] In addition, this invention is mainly used to increase the accuracy of user-input query results. Therefore, in addition to querying names, it can also be applied to various query services in user voice interaction scenarios, such as finding organizations and departments.

[0104] This invention provides a method for finding people in a contact list by converting user text to text and extracting keywords to identify a keyword set. This enables basic contact search operations and yields a result set. Then, by identifying user information and that of recommended individuals, a user knowledge graph and various recommended knowledge graphs are generated. These knowledge graphs are clustered to obtain clustering scores, which represent the proximity of the relationship between each recommended individual and the user, thus prioritizing user recommendations. Furthermore, the result set incorporates fuzzy search to avoid keyword errors caused by speech-to-text conversion. Therefore, the result set may contain multiple names. Finally, the recommended individuals are sorted based on name character length and clustering relationships to obtain a ranking result, enabling personalized search results and more accurate results. Therefore, this invention provides a contact list-based contact search recommendation method that significantly improves the accuracy of contact search.

[0105] like Figure 5 The diagram shown is a functional block diagram of a contact list-based person-finding and recommendation device provided in an embodiment of the present invention.

[0106] The address book-based person-finding recommendation device 100 of this invention can be installed in an electronic device. Depending on the functions implemented, the address book-based person-finding recommendation device 100 may include a text acquisition module 101, a keyword-based person-finding module 102, a knowledge graph construction module 103, and a clustering query module 104. The module described in this invention can also be called a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and are stored in the memory of the electronic device.

[0107] In this embodiment, the functions of each module / unit are as follows:

[0108] The text acquisition module 101 is used to acquire voice data input by the user and convert the voice data into text to obtain text sentences;

[0109] The keyword-based person-finding module 102 is used to extract keywords from the text statement to obtain a keyword set, and to query a pre-built address book database based on the keyword set to obtain a set of person-finding results.

[0110] The knowledge graph construction module 103 is used to obtain user information and construct a user knowledge graph using the user information, and to obtain target information of each recommended person in the missing person result set and construct a recommendation knowledge graph corresponding to each recommended person based on the target information.

[0111] The clustering query module 104 is used to perform text clustering on the user knowledge graph and each of the recommendation knowledge graphs to obtain the clustering scores corresponding to the user and each of the recommended individuals, and to perform a sorting operation on each of the recommended individuals based on the name character length and clustering relationship according to the preset recommendation sorting strategy and the clustering scores to obtain the recommendation sorting result.

[0112] In detail, the modules in the address book-based person-finding recommendation device 100 described in this application embodiment adopt the same usage as described above. Figures 1 to 5 The method used is the same as the address book-based person-finding recommendation method described above, and it can produce the same technical effect, so it will not be repeated here.

[0113] like Figure 6 The diagram shown is a schematic representation of the structure of an electronic device 1 that implements a contact list-based person-finding recommendation method according to an embodiment of the present invention.

[0114] The electronic device 1 may include a processor 10, a memory 11, a communication bus 12, and a communication interface 13. It may also include a computer program stored in the memory 11 and capable of running on the processor 10, such as a contact list-based missing person recommendation program.

[0115] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the electronic device 1, connecting various components of the electronic device through various interfaces and lines. It executes programs or modules stored in the memory 11 (e.g., executing a contact list-based missing person recommendation program) and calls data stored in the memory 11 to perform various functions of the electronic device and process data.

[0116] The memory 11 includes at least one type of readable storage medium, including flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of an electronic device, such as a portable hard drive. In other embodiments, the memory 11 can be an external storage device of the electronic device, such as a plug-in portable hard drive, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, etc. Furthermore, the memory 11 can include both internal and external storage units of the electronic device. The memory 11 can be used not only to store application software and various types of data installed on the electronic device, such as the code of a contact-based missing person recommendation program, but also to temporarily store data that has been output or will be output.

[0117] The communication bus 12 can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This bus can be divided into an address bus, a data bus, a control bus, etc. The bus is configured to enable communication between the memory 11 and at least one processor 10, etc.

[0118] The communication interface 13 is used for communication between the electronic device 1 and other devices, including a network interface and a user interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a Wi-Fi interface, Bluetooth interface, etc.), typically used to establish communication connections between the electronic device and other electronic devices. The user interface may be a display, an input unit (such as a keyboard), or, optionally, a standard wired or wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device and to display a visual user interface.

[0119] Figure 6 Only electronic devices with components are shown; those skilled in the art will understand that... Figure 6 The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown, or combine certain components, or have different component arrangements.

[0120] For example, although not shown, the electronic device 1 may also include a power supply (such as a battery) to power various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power supplies, recharging devices, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device 1 may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0121] It should be understood that the embodiments described are for illustrative purposes only and are not limited to this structure in the scope of the patent application.

[0122] The address book-based missing person recommendation program stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When run in the processor 10, it can achieve the following:

[0123] Acquire user-input voice data and convert the voice data to text to obtain text statements;

[0124] Keyword extraction is performed on the text statement to obtain a keyword set, and the pre-built address book database is queried based on the keyword set to obtain a set of missing person results;

[0125] Obtain user information and use the user information to construct a user knowledge graph; obtain target information of each recommended person in the missing person result set and construct a recommendation knowledge graph corresponding to each recommended person based on the target information.

[0126] The user knowledge graph and each of the recommendation knowledge graphs are subjected to text clustering to obtain the clustering scores corresponding to the user and each of the recommended individuals.

[0127] Based on the preset recommendation ranking strategy and the clustering score, the recommended individuals are ranked according to their name character length and clustering relationship to obtain the recommendation ranking result.

[0128] Specifically, the specific implementation method of the processor 10 for the above instructions can be referred to the description of the relevant steps in the corresponding embodiment of the accompanying drawings, and will not be repeated here.

[0129] Furthermore, if the modules / units integrated in the electronic device 1 are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0130] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor of an electronic device, can perform the following:

[0131] Acquire user-input voice data and convert the voice data to text to obtain text statements;

[0132] Keyword extraction is performed on the text statement to obtain a keyword set, and the pre-built address book database is queried based on the keyword set to obtain a set of missing person results;

[0133] Obtain user information and use the user information to construct a user knowledge graph; obtain target information of each recommended person in the missing person result set and construct a recommendation knowledge graph corresponding to each recommended person based on the target information.

[0134] The user knowledge graph and each of the recommendation knowledge graphs are subjected to text clustering to obtain the clustering scores corresponding to the user and each of the recommended individuals.

[0135] Based on the preset recommendation ranking strategy and the clustering score, the recommended individuals are ranked according to their name character length and clustering relationship to obtain the recommendation ranking result.

[0136] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0137] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0138] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0139] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0140] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within the invention. No appended diagram markings in the claims should be construed as limiting the scope of the claims.

[0141] The blockchain referred to in this invention is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked together using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. A blockchain can include an underlying blockchain platform, a platform product service layer, and an application service layer.

[0142] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0143] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices recited in a system claim may also be implemented by a single unit or device through software or hardware. The terms "first," "second," etc., are used to indicate names and do not indicate any specific order.

[0144] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for recommending a person to find based on a contact list, characterized in that, The method includes: Acquire user-input voice data and convert the voice data to text to obtain text statements; Keyword extraction is performed on the text statement to obtain a keyword set, and the pre-built address book database is queried based on the keyword set to obtain a set of missing person results; Obtain user information and use the user information to construct a user knowledge graph; obtain target information of each recommended person in the missing person result set and construct a recommendation knowledge graph corresponding to each recommended person based on the target information. Using the cosine similarity algorithm, the similarity between the user knowledge graph and each recommendation knowledge graph is calculated based on multiple entity types to obtain the similarity scores of multiple entity types between the user and each recommendation person. The knowledge graph is weighted according to preset weight coefficients, and the similarity scores of the multiple entity types are weighted according to the weight configuration relationship to obtain the clustering scores of the user and each recommended person. Based on the preset recommendation ranking strategy and the clustering score, the recommended individuals are ranked according to their name character length and clustering relationship to obtain the recommendation ranking result.

2. The contact-based person-finding and recommendation method as described in claim 1, characterized in that, The step of constructing a user knowledge graph using the user information includes: The feature extraction network of the pre-built graph construction model is used to perform entity extraction, relation recognition and attribute recognition operations on the user information to obtain entity set, relation set and attribute set; The graph construction network of the model is constructed using the graph, and entity linking operations are performed according to the correspondence between the entity set, the relation set, and the attribute set to obtain a knowledge representation relation set; According to a preset knowledge fusion strategy, ambiguity is eliminated for each knowledge expression relation in the knowledge expression relation set to obtain a fact expression relation set. Based on the preset conceptual framework, hierarchical identification and entity classification operations are performed on each entity in the fact expression relationship set. Based on the hierarchical identification results and entity classification results, the fact expression relationship set is arranged in sequence to obtain the user knowledge graph.

3. The address book-based person search recommendation method according to claim 1, wherein The step of sorting each recommended person based on name character length and clustering relationship according to a preset recommendation ranking strategy and the clustering score to obtain the recommendation ranking result includes: Obtain the name length of each recommended person to get the name character length; According to a preset recommendation arrangement strategy, the name character length is grouped to obtain multiple name recommendation groups, and the name recommendation groups are arranged in order from longest to shortest name character length. Based on the cluster scores of each recommended person from highest to lowest, the recommended groups of names are arranged within each group to obtain the recommendation ranking result.

4. The address book-based person search recommendation method according to claim 1, wherein The process of extracting keywords from the text statement to obtain a keyword set includes: The text statement is segmented using a word segmentation tool to obtain a set of segmentation results. The word segmentation result set is cleaned by deleting stop words from the preset stop word dictionary. The word sense category of each word in the clean word segmentation result set is identified by using a pre-trained target entity recognition model to obtain the category recognition result, and word segments of a preset type are extracted from the category recognition result to obtain a keyword set.

5. The address book based people search recommendation method of claim 1, wherein, The step of querying a pre-built address book database based on the keyword set to obtain a set of search results includes: Based on the keyword set, the user's historical communication records are precisely matched to obtain a priority result set; A fuzzy search is performed on the target type strings in the keyword set to the address book database, and an exact match is performed on the keyword characters other than the target type strings in the keyword set to the address book database to obtain a preliminary set of missing person results. Based on the priority result set, a portion of the search results in the primary search result set are prioritized to obtain the search result set.

6. The address book based people search recommendation method of claim 1, wherein, After performing a sorting operation on each recommended user based on name character length and clustering relationship to obtain the recommendation ranking result, the method further includes: When a user selects a target person from the recommended ranking results, the target person and the user are stored in a preset query response count table; By using a pre-defined data extraction whitelist, the social relationships of target users across multiple target communication applications can be obtained. The user knowledge graph is dynamically updated based on the social relationships and the query response count table.

7. A contact-based person search recommendation device for implementing the contact-based person search recommendation method according to any one of claims 1 to 6, characterized by The device includes: The text acquisition module is used to acquire the voice data input by the user and convert the voice data into text to obtain text sentences; The keyword-based person search module is used to extract keywords from the text statement to obtain a keyword set, and then query a pre-built address book database based on the keyword set to obtain a set of person search results. The knowledge graph construction module is used to obtain user information and construct a user knowledge graph using the user information, and to obtain target information of each recommended person in the missing person result set and construct a recommendation knowledge graph corresponding to each recommended person based on the target information. The clustering query module is used to perform text clustering on the user knowledge graph and each of the recommendation knowledge graphs to obtain the clustering scores corresponding to the user and each of the recommended individuals, and to perform a sorting operation on each of the recommended individuals based on the name character length and clustering relationship according to the preset recommendation sorting strategy and the clustering scores to obtain the recommendation sorting results.

8. An electronic device, characterized in that, The electronic device includes: At least one processor; and, A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the address book-based person-finding recommendation method as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the address book-based person-finding recommendation method as described in any one of claims 1 to 6.

Images