Method and system for constructing record knowledge graph

Abstract

The embodiment of the invention relates to the technical field of information, in particular to a record knowledge graph construction method and system. Comprising the following steps: reading a record, extracting a plurality of keywords of the record, and matching a knowledge structure template according to the keywords; generating a plurality of knowledge according to the record and the knowledge structure template; according to events related to the knowledge, if matched evidence on the case exists, adding information of the evidence on the case to the knowledge; trying to obtain a corresponding public area video monitoring fragment according to the time and the place involved by the knowledge; if the corresponding public area video monitoring clip exists, identifying an event in the public area video monitoring clip, and recording the event as an identification event; comparing the identification event with an event related to the knowledge to obtain a matching degree, and adding the public area video monitoring clip and the matching degree into the knowledge; and representing all the obtained knowledge by using a graph to obtain a record knowledge graph.

Description

Technical Field

[0001] Multiple embodiments of this specification relate to the field of information technology, and specifically to a method and system for constructing a transcript knowledge graph. Background Art

[0002] With the rapid advancement of information technology and the increasing demand for public safety in society, the massive amounts of data generated when handling various cases have become increasingly important. These data are not only the key to case handling, but also the cornerstone of ensuring social harmony and stability. As one of the core sources of case information, transcripts contain a large amount of useful information, covering a variety of information such as time, location, relationships between people, and motives. However, traditional transcript analysis methods are highly dependent on manual participation, which not only consumes a lot of human resources, but also the analysis process is very time-consuming. Staff members need to read and understand each transcript word by word. This process is not only inefficient, but also prone to errors due to fatigue or distraction, which seriously affects the speed and quality of case handling. This makes it difficult to effectively extract valuable relevant clues from the complex information. When dealing with complex cases, this method that relies on manpower is difficult to reveal the deep connections hidden in big data, which poses a challenge to the rapid and efficient resolution of cases. For this reason, it is necessary to improve the technology of transcript processing. Summary of the invention

[0003] Multiple embodiments of this specification describe a method and system for constructing a transcript knowledge graph.

[0004] In a first aspect, the embodiments of this specification provide a method for constructing a transcript knowledge graph, comprising the steps of: Read the transcript, extract several keywords from the transcript, and match the knowledge structure template according to the keywords; the knowledge structure template records the fields contained in the knowledge, Generate a number of pieces of knowledge according to the transcript and the knowledge structure template; According to the events involved in the knowledge, try to match the evidence on file, and if there is matching evidence on file, add the information of the evidence on file to the knowledge; According to the time and place involved in the knowledge, try to obtain the corresponding public area video surveillance clips; If there is a corresponding public area video surveillance clip, identifying an event in the public area video surveillance clip, and recording it as an identified event; Comparing the identified event with the event involved in the knowledge to obtain a matching degree, and adding the public area video surveillance clip and the matching degree to the knowledge; All acquired knowledge is represented using a graph to obtain a knowledge graph of the transcript.

[0005] In a second aspect, the embodiments of this specification provide a system for constructing a transcript knowledge graph, including: The reading module reads the transcript, extracts a number of keywords from the transcript, and matches the knowledge structure template according to the keywords; the knowledge structure template records the fields contained in the knowledge, A knowledge module generates a plurality of pieces of knowledge according to the transcript and the knowledge structure template; A first matching module attempts to match the on-record evidence according to the event involved in the knowledge, and if there is matching on-record evidence, adds the information of the on-record evidence to the knowledge; The second matching module attempts to obtain corresponding public area video surveillance clips according to the time and location involved in the knowledge; The recognition module recognizes an event in the public area video surveillance segment if there is a corresponding public area video surveillance segment, and records it as a recognition event; A matching module, which compares the identified event with the event involved in the knowledge, obtains a matching degree, and adds the public area video surveillance clip and the matching degree to the knowledge; The construction module represents all the acquired knowledge using a graph to obtain the transcript knowledge graph.

[0006] In a third aspect, an embodiment of this specification provides an electronic device, including a processor and a memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the method described in any one of the above aspects.

[0007] In a fourth aspect, an embodiment of the present specification provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method described in any of the above aspects.

[0008] In a fifth aspect, the embodiments of this specification provide a computer program product, including a computer program, which implements the method described in any of the above aspects when executed by a processor.

[0009] The beneficial effects brought by the technical solutions provided by some embodiments of this specification include at least: In multiple embodiments of this specification, the provided transcript knowledge graph construction method can extract knowledge from the transcript based on keywords and knowledge structure templates, automatically construct a knowledge graph based on the acquired knowledge, and match the evidence in the case and the video surveillance clips in public areas, providing a comprehensive reference for case handlers, which helps to improve case handling efficiency. It can combine multiple transcripts to construct a knowledge graph, which is suitable for interpreting transcripts in complex cases with a large number of transcripts.

[0010] Other features and advantages of the various embodiments of the present specification will be further disclosed in the following detailed description and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of this specification, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this specification. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0012] Figure 1 A schematic diagram of the application architecture of the transcript knowledge graph construction method provided in the embodiments of this specification.

[0013] Figure 2 A flowchart of the method for constructing a transcript knowledge graph provided in an embodiment of this specification.

[0014] Figure 3 A flowchart of a method for extracting transcript keywords provided in an embodiment of this specification.

[0015] Figure 4 A schematic diagram of the flow chart of the method for presetting a knowledge structure template provided in an embodiment of this specification.

[0016] Figure 5 A schematic flow chart of a method for obtaining public area video surveillance clips provided in an embodiment of this specification.

[0017] Figure 6 A schematic diagram of the flow of a method for obtaining relevant public areas provided in an embodiment of this specification.

[0018] Figure 7 A schematic diagram of a system for constructing a transcript knowledge graph according to an embodiment of this specification.

[0019] Figure 8 A schematic diagram of an electronic device provided in an embodiment of this specification. DETAILED DESCRIPTION

[0020] The following is an explanation and description of the technical solutions of the embodiments of this specification in conjunction with the drawings of the embodiments of this specification, but the following embodiments are only preferred embodiments of this specification, not all. Based on the embodiments in the implementation mode, other embodiments obtained by those skilled in the art without creative work are all within the scope of protection of this specification.

[0021] The terms "first", "second", "third", etc. in the description and claims of this specification and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0022] In the following description, terms such as "inside", "outside", "up", "down", "left", "right", etc. that indicate directions or positional relationships are only used to facilitate the description of the embodiments and simplify the description, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as a limitation of this specification.

[0023] The data involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection of relevant data complies with the relevant laws, regulations and standards of relevant countries and regions.

[0024] Before describing the technical solution in this specification, the application scenarios and related technologies of the technical solution are introduced.

[0025] The transcript 10 is a written material that records the evidence collected during the investigation of a case. It includes but is not limited to the record of the questioning of witnesses and victims, and the statement of the confession of the suspect. The transcript 10 is an important part of the judicial process, and it provides first-hand evidence for the trial of the case. The application scope of the transcript 10 is relatively wide. The transcript 10 is required in the following situations. When the case is initially reported, when an alarm is received or a potential illegal act is discovered, it is necessary to record the statement of the reporter. During the on-site investigation, during the on-site investigation, it is necessary to record information such as the on-site situation, the location of physical evidence, and the environmental conditions. When questioning witnesses and suspects, during the investigation, it is necessary to record the information provided by the witnesses and the content of the suspect's confession. During the interrogation process, during the formal interrogation, it is necessary to record the suspect's response to the accusation and his defense opinion. In the mediation agreement, during the mediation of civil disputes, after the two parties reach an agreement, it is necessary to record the mediation results. It can be seen that the transcript 10 runs through the entire process of case handling and is an important carrier for recording case materials.

[0026] The transcript usually includes: the time and place of the record, the basic information of the person being questioned (such as name, age, occupation, etc.), case-related questions and their answers, the description and handling of physical evidence, a detailed description of the scene, and the part signed and confirmed by both parties. The transcript converts oral narration into written text, fixes the evidence, and prevents subsequent forgetfulness or dispute. The records and information provided help law enforcement officers understand the case and provide reference for the subsequent investigation direction. It helps to clarify the responsibilities of all parties and provide support for possible legal proceedings. The existence of the transcript also plays a certain supervisory role to prevent improper behavior in the law enforcement process. The accuracy and completeness of the transcript are directly related to whether the case can be handled fairly and reasonably.

[0027] In this specification, in order to assist in the interpretation of transcripts, a technology for constructing a transcript knowledge graph is provided. Figure 1 The technical solution provided in this specification runs in the server 20, and the server 20 can read multiple transcripts 10. The server 20 can be connected to the terminal device 30, and the terminal device 30 can select which transcripts 10 to build the knowledge graph based on, and view the knowledge graph that has been built. The terminal device 30 runs an operation interface for selecting and uploading the transcript 10, as well as an interface for viewing the knowledge graph. The server 20 uses a database system to store the knowledge graph.

[0028] The interface for viewing the knowledge graph can be run on the terminal device 30 in the form of a browser, or in the form of an independent application, etc. The specific display form of the terminal device 30 is not limited here. The server 20 involved in this application can be an independent physical server 20, or a server 20 cluster or distributed system composed of multiple physical servers 20, or a cloud server 20 that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms. The terminal device 30 can be a smart phone, a tablet computer, a laptop, a PDA, a personal computer, a smart speaker, a smart TV, a smart watch, a car device, a wearable device, etc., but is not limited to this. The terminal device 30 and the server 20 can be directly or indirectly connected by wired or wireless communication, and this application is not limited here. The number of servers 20 and terminal devices 30 is also not limited. The solution provided in the present application can be completed independently by the terminal device 30, can be completed independently by the server 20, or can be completed by the terminal device 30 and the server 20 in cooperation. The present application does not make any specific limitation on this.

[0029] In short, a database is like an electronic filing cabinet, a place for storing electronic files, where users can perform operations such as adding, retrieving, modifying and deleting data. A database refers to a collection of data organized in a specific way, which can be shared by multiple users, has minimal redundancy and is relatively independent of application software. A database management system (DBMS) is a computer software system used to manage databases, usually with functions such as storage, retrieval, security protection and backup. DBMS can be divided into many types according to different standards. For example, according to the database models they support, there are relational databases and databases based on extensible markup language (XML); according to the computing platforms they support, there are databases suitable for server clusters and mobile devices; according to the query languages ​​they support, there are databases that support structured query language (SQL) and XQuery; or according to their performance focus, such as databases that pursue maximum capacity or highest operating speed. In addition, they can be classified according to other characteristics. It is worth noting that some DBMS can cross multiple classification standards, such as supporting multiple query languages ​​at the same time.

[0030] Since this application involves some professional terms, these professional terms will be introduced below.

[0031] record A transcript is a written record compiled by law enforcement officers or authorized personnel based on the content of case investigation, interrogation and other activities during legal proceedings. It is a common form of document in the judicial and administrative law enforcement process, used to record relevant facts, evidence and statements of the parties. The transcript includes the object of the record, the content of the record, the time and place of the record, the recorder, and the confirmation signature.

[0032] Knowledge Graph Knowledge graph is a data structure that organizes information into entities and relationships between entities. It is often used in semantic networks and databases. Knowledge graphs use graph theory to express the connections between things, enabling machines to understand and process this information, and thus provide smarter services. Knowledge graphs include: nodes, which represent entities such as people, places, events, and concepts; edges, which represent the relationships between entities, such as "friendship" and "located at"; and properties, which are attached to nodes or edges to provide additional information.

[0033] Public Areas Public areas usually refer to places that are open to the public or for public use. Where permitted by law and policy, surveillance cameras may be installed in these places to ensure public safety and public order. Such areas generally include but are not limited to streets, squares, parks, public transportation vehicles and their stations, government offices, schools, hospitals, shopping malls, and any other non-privately owned and open spaces to the public. The installation of surveillance cameras must comply with relevant laws and regulations. The surveillance cameras involved in this manual are all surveillance cameras installed in accordance with relevant regulations.

[0034] Map Data Map data refers to GIS data, which is Geographic Information System (GIS). GIS data combines geographic location information with various statistical data, and uses computer hardware, software and network technology to realize the collection, storage, processing, analysis, display and application of geographic spatial data. GIS uses maps and other graphics to display data, allowing users to understand complex relationships, patterns and processes more intuitively. Example

[0035] This manual first provides a method for constructing a transcript knowledge graph. Figure 2 , including the steps of: Step S101) Read the transcript 10, extract several keywords from the transcript 10, and match the knowledge structure template according to the keywords. The knowledge structure template records the fields contained in the knowledge, identifies the fields to which the keywords belong, and obtains several corresponding knowledge by comparing the fields to which the keywords belong with the fields contained in the knowledge recorded in the knowledge structure template. The knowledge structure template with the most corresponding knowledge is the knowledge structure template matched by the extracted keywords. Exemplarily, the keywords extracted from the transcript 10 are: prying open the door lock, rummaging, cash, gold jewelry, and lookout. The knowledge structure template records the fields of knowledge related to theft. The fields that should be recorded for the knowledge related to theft are: the time of the case, the location of the case, details of the stolen items, witness information, video surveillance records, suspect characteristics, theft modus operandi, information of the person who reported the case, on-site conditions, and the time of reporting the case. Among them, the field corresponding to the keyword prying open the door lock is the theft modus operandi, the fields corresponding to rummaging and lookout are also the theft modus operandi, and the fields corresponding to cash and gold jewelry are the details of the stolen items.

[0036] Step S102) Generate a plurality of pieces of knowledge according to the transcript 10 and the knowledge structure template.

[0037] As in the above example, continue to read and extract keywords from transcript 10, and combine with the knowledge structure template to obtain the following knowledge: <time of occurrence of the case, 10 p.m. on August 13>, <location of the case, Room 1101, Unit 1, Building 1, Jinjin Community>, <details of stolen items, cash>, <details of stolen items, gold jewelry>, <information of witnesses, NULL>, <video surveillance record, Video0813223101302589>, <characteristics of suspects, NULL>, <method of theft, prying open the door>, <method of theft, rummaging>, <method of theft, lookout>, <information of the person who reported the case, NULL>, <conditions on the scene, traces of drawers being rummaged through>, <time of reporting the case, 9 a.m. on August 14>.

[0038] Step S103) According to the events involved in the knowledge, try to match the evidence on file. If there is matching evidence on file, add the information of the evidence on file to the knowledge.

[0039] The registered photos of the stolen cash and gold jewelry are evidence in the case and are added to the knowledge <Details of stolen items, cash> and knowledge <Details of stolen items, gold jewelry>. The photos of the theft scene are also evidence in the case and are added to the knowledge <Situation on the scene, traces of drawers being searched>. The photos of the pried open door lock and related inspection information are also evidence in the case and are added to the knowledge <Theft modus operandi, prying open the door lock>.

[0040] Step S104) According to the time and place involved in the knowledge, try to obtain the corresponding public area video surveillance clip.

[0041] Video surveillance clip Video0813223101302589 is a video clip shot at 10 pm on August 13th, which is a camera installed at the entrance and exit of the unit where the theft occurred in Jinjin Community. It captures the process of the suspect entering the unit. According to the knowledge <Case time, August 13th 10 pm> and the knowledge <Case location, Jinjin Community 1st Building 1st Unit 1101>, the corresponding public areas can be obtained as the entrance and exit of Unit 1 in Building 1 and the entrance and exit of Jinjin Community. Priority is given to inspecting the surveillance videos of the unit entrances and exits, and obtaining the surveillance videos that captured the suspect entering Unit 1, that is, obtaining the video surveillance clip Video0813223101302589.

[0042] Step S105) If there is a corresponding public area video surveillance segment, an event in the public area video surveillance segment is identified and recorded as an identified event.

[0043] Identify the event in the video surveillance clip Video0813223101302589, the event is XX person entering and leaving unit 1. When the event of the suspect entering and leaving unit 1 is identified, it is recorded as a recognition event.

[0044] Step S106) Compare the identified event with the event involved in the knowledge to obtain a matching degree, and add the public area video surveillance clip and the matching degree to the knowledge.

[0045] The identified event "suspect enters and exits unit 1" is compared with the event involved in the knowledge "suspect burglarizes a house", and the matching degree is 93%. At this time, the public area video surveillance clip Video0813223101302589 and the matching degree of 93% are added to the knowledge <time of the case, 10 pm on August 13> and the knowledge <location of the case, room 1101, unit 1, building 1, Jinjin community>.

[0046] Step S107) All the acquired knowledge is represented by a graph to obtain a transcript knowledge graph.

[0047] This embodiment uses triples to represent the knowledge graph. The triple is "source-relation-target", and the triple structure corresponds to the nodes and edges in the graph. The nodes correspond to the source and target, and the edges correspond to the relation. Specifically, the source node is unified as Zhang San's burglary case, the target node corresponds to the field of knowledge, and the relation corresponds to the content of knowledge. Exemplarily, the knowledge <suspect, Zhang San> is converted into a triple as {"Zhang San's burglary case", "Zhang San", "suspect"}, and the knowledge <time of the case, 10 o'clock on the evening of August 13> is converted into a triple as {"Zhang San's burglary case", "10 o'clock on the evening of August 13", "time of the case"}. By analogy, all knowledge can be converted into triples. Then it is stored in the database in the form of triples. And the knowledge graph is constructed using nodes and edges.

[0048] For a method of extracting several keywords from the transcript 10, please refer to the attached Figure 3 ,include: Step S201) skipping the first several paragraphs of the transcript 10 according to several preset first-category reference words; Step S202) continuously reading the transcript 10 by paragraph, and extracting paragraph keywords of the paragraph; Step S203) calculating the relevance between the subsequent paragraph keyword and the previous paragraph keyword obtained, and taking the maximum value of the relevance as the final value of the relevance; Step S204) The paragraph keywords whose final relevance values ​​are less than a preset threshold are deleted, and the keywords of the transcript 10 are obtained based on the remaining paragraph keywords.

[0049] For example, the contents of the transcript are as follows: Time: 11:30 on September 28, 2024 to 13:00 on September 28, 2024 Location: Interrogation Room of XX Team, XX Bureau, XX District, XX City Inquiring persons: Li XX, Zhang XX Recorder: Wang XX Person being questioned: Xiao Liu (pseudonym) Inquiry content: Interviewer: Please describe in detail your role in the plan and how you worked with Kate to crack the security system.

[0050] Interviewee: I am the technical expert of the team, mainly responsible for cracking the security system. Kate is a hacker from abroad, she is responsible for remote operation and providing technical support. We have clearly divided the work in the planning stage.

[0051] The preset reference words include but are not limited to "police force", "interrogation room", "interrogator", and "recorder". These paragraphs are formatted content and need to be skipped. It is worth noting that different types of transcripts 10 have different contents to be skipped. For example, the inspection transcript does not have an interrogator, but an investigator. It is necessary to set a reference word according to different transcript types. It is also possible to collect a reference word of all types of transcripts 10 into one set, and compare the specific transcript 10 read with the set.

[0052] Then continue to read the transcript 10 by paragraph. In the previous example, the transcript will be read from the inquiry content, and the paragraph keywords of the first paragraph are "description", "role", "Kate", "cooperation", "cracking", and "security system".

[0053] Then read the second paragraph, and obtain the paragraph keywords of the second paragraph, including "technical expert", "crack", "security system", "Kate", "foreign", "hacker", "remote operation", "technical support", "plan", and "division of labor". Among them, the maximum value of the correlation between the paragraph keyword "technical expert" in the second paragraph and the paragraph keyword in the first paragraph is the correlation with "crack". The specific correlation between "technical expert" and "crack" is 80%. The maximum value of the correlation between the paragraph keyword "crack" in the second paragraph and the paragraph keyword in the first paragraph is the correlation with "crack", that is, 100%. The maximum value of the correlation between the paragraph keyword "foreign" in the second paragraph and the paragraph keyword in the first paragraph is the correlation with "role", that is, 40%. The maximum value of the correlation between the paragraph keyword "plan" in the second paragraph and the paragraph keyword in the first paragraph is the correlation with "cooperation", that is, 80%. When the preset threshold is 60%, the paragraph keyword "foreign" in the second paragraph will be deleted. The subsequent paragraphs are deduced by analogy. Finally, according to the remaining paragraph keywords, the keywords of transcript 10 are obtained. It is possible to filter out low-weight keywords in the transcript 10, thereby improving the accuracy of the transcript keywords.

[0054] The method of matching the knowledge structure template according to the keyword includes: Reading a preset knowledge structure template, wherein the knowledge structure template is associated with a plurality of matching words; The matching degree between the keyword and the matching word is calculated, and a matching knowledge structure template is obtained according to the matching degree.

[0055] For example, the matching words of the knowledge structure template of the transcript 10 used to record theft cases are: "pry", "search", "lookout", "crack", "security system", and "anti-theft system". By matching the keywords extracted from the transcript 10 with these matching words, it can be known that the keywords extracted from the transcript 10 can match the knowledge structure template of the transcript 10 used to record theft cases.

[0056] Please see attached Figure 4 , the methods for presetting the knowledge structure template include: Step S301) receiving manually set key information tables and sample transcripts of various types of cases; Step S302) generating a knowledge structure template according to the key point information table; Step S303) Extract and obtain a number of matching words based on the sample transcript of the case type.

[0057] For example, the key information of theft cases set manually includes the method of committing the crime, stolen items, theft time, suspect, stolen person, and stolen address. The knowledge structure template is generated based on the key information, specifically, the key information is used as fields, each field is given a null value, and finally the knowledge structure template is formed.

[0058] On the other hand, in another embodiment, the method of matching the evidence on file includes: Reading a pre-configured reference evidence table, wherein the reference evidence table records several evidence types and evidence descriptions corresponding to the event; Obtaining several types of evidence and evidence descriptions corresponding to the event according to the reference evidence table; The evidence on file is compared with the evidence type. If the types match, the knowledge is matched with the evidence description. If they match, it is determined that the evidence on file matches.

[0059] For example, please see the attached Figure 5 , the types of evidence for the "lost property" incident include: "lost scene" and "stolen property". The evidence of "lost scene" is described as the traces left at the scene of lost property, such as traces of rummaging, traces of damaged packaging, etc. The evidence of "stolen property" is described as the cash and items that have been recovered.

[0060] On the other hand, in another embodiment, the method for obtaining the corresponding public area video surveillance segment includes: Step S401) obtaining relevant public areas of the locations involved in the knowledge; Step S402) searching for corresponding surveillance cameras according to the relevant area; Step S403) If there is a corresponding surveillance camera, then obtain a public area video surveillance clip according to the time involved in the knowledge.

[0061] If the knowledge involves a location, for example, the knowledge <location of the case, Jinjin Community 1 Building 1 Unit 1 Room 1101> means that the location where the theft occurred is Jinjin Community 1 Building 1 Unit 1 Room 1101. According to "Jinjin Community 1 Building 1 Unit 1 Room 1101", the relevant public areas can be obtained, including the community entrance and exit, within the community, unit entrance and exit, elevator, and 11th floor corridor. According to these relevant areas, the corresponding surveillance cameras are found, and it is found that the community entrance and exit, unit entrance and exit, and elevator are monitored, while the community and 11th floor corridor are not monitored. According to the knowledge <time of the case, August 13 at 10 pm>, the surveillance of the community entrance and exit, unit entrance and exit, and elevator for 1 hour before and after 10 pm on August 13 is obtained as the public area video surveillance clip.

[0062] On the other hand, in another embodiment, please refer to the attached Figure 6, the method of obtaining the relevant public area of ​​the place involved in the knowledge includes: Step S501) obtaining the location coordinates of the place involved in the knowledge; Step S502) reading pre-configured map data to obtain a public area near the location; Step S503) It is determined whether the nearby public area can reach the location, and the public area that can be reached is used as the relevant public area of ​​the location involved in the knowledge.

[0063] For example, the knowledge <location of the case, Room 1101, Unit 1, Building 1, Jinjin Community> means that the location where the theft occurred is Room 1101, Unit 1, Building 1, Jinjin Community. Read the accessed map data, i.e. GIS data, which areas are marked as public areas in the map data. Or identify map objects based on map data, and identify which are public areas based on map objects. For example, by identifying elements such as roads, walls, access control, doors, elevators, entrances and exits, the map data can be identified and interpreted to obtain the public areas near the location involved in the knowledge. The identification of map objects can be manually annotated to form sample data, which can be implemented by machine learning models. Or other technologies disclosed in the art can be used. After identifying map elements such as roads, access control, entrances and exits, elevators, doors, walls, etc., it can be determined that a certain public area can reach the location. Since the scope of the case should be appropriate and not too large, in some implementations, the vicinity can be limited to 10 meters or other close values. In this way, the entrance and exit of the community can be excluded from the vicinity of the case. Because the flow of people at the entrance and exit of the community is too large, the reference value is not high. However, the unit entrances and elevators are still within the vicinity of where the incident occurred.

[0064] On the other hand, this specification provides a system for constructing a transcript knowledge graph, see Appendix 7, including: The reading module 100 reads the transcript 10, extracts a number of keywords from the transcript 10, and matches the knowledge structure template according to the keywords; the knowledge structure template records the fields included in the knowledge, The knowledge module 200 generates a plurality of knowledge according to the transcript 10 and the knowledge structure template; A first matching module 300 attempts to match the on-record evidence according to the event involved in the knowledge, and if there is matching on-record evidence, adds the information of the on-record evidence to the knowledge; The second matching module 400 attempts to obtain a corresponding public area video surveillance segment according to the time and location involved in the knowledge; The identification module 500 identifies an event in the public area video surveillance segment if there is a corresponding public area video surveillance segment, and records it as an identified event; A matching module 600 compares the identified event with the event involved in the knowledge to obtain a matching degree, and adds the public area video surveillance clip and the matching degree to the knowledge; The construction module 700 represents all the acquired knowledge using a graph to obtain a transcript knowledge graph.

[0065] See also Figure 8 A schematic diagram of the structure of an electronic device provided in an embodiment of this specification is shown.

[0066] like Figure 8 As shown, the electronic device 1100 may include: at least one processor 1101, at least one network interface 1104, a user interface 1103, a memory 1105 and at least one communication bus 1102. Among them, the communication bus 1102 can be used to realize the connection and communication of the above-mentioned components. Among them, the user interface 1103 may include a button, and the optional user interface may also include a standard wired interface and a wireless interface. Among them, the network interface 1104 may include but is not limited to a Bluetooth module, an NFC module, a Wi-Fi module, etc. Among them, the processor 1101 may include one or more processing cores. The processor 1101 uses various interfaces and lines to connect various parts in the entire electronic device 1100, and executes various functions and processes data of the routing device 1100 by running or executing instructions, programs, code sets or instruction sets stored in the memory 1105, and calling data stored in the memory 1105. Optionally, the processor 1101 can be implemented in at least one hardware form of DSP, FPGA, and PLA. The processor 1101 may integrate one or a combination of CPU, GPU and modem, etc. Among them, the CPU mainly processes the operating system, user interface and application programs, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communication.

[0067] It is understandable that the above-mentioned modem may not be integrated into the processor 1101, but may be implemented by a separate chip.

[0068] Among them, the memory 1105 may include RAM or ROM. Optionally, the memory 1105 includes a non-transitory computer-readable medium. The memory 1105 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 1105 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned various method embodiments, etc.; the data storage area may store data involved in the above-mentioned various method embodiments, etc. The memory 1105 may also be at least one storage device located away from the aforementioned processor 1101. The memory 1105 as a computer storage medium may include an operating system, a network communication module, a user interface module and an application. The processor 1101 may be used to call the application stored in the memory 1105 and execute the methods in the above-mentioned multiple embodiments.

[0069] The embodiments of this specification also provide a computer-readable storage medium, which stores instructions, and when the instructions are executed on a computer or a processor, the computer or the processor executes the multiple steps in the above embodiments. If the components of the above electronic device are implemented in the form of software functional units and sold or used as independent products, they can be stored in the computer-readable storage medium.

[0070] The embodiments of this specification also provide a computer program product, including a computer program, which implements multiple steps in the above embodiments when executed by a processor.

[0071] In the absence of conflict, the technical features in this embodiment and implementation scheme can be combined arbitrarily.

[0072] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes a plurality of computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of this specification is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions may be transmitted from a website site, a computer, a server 20 or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (Digital Subscriber Line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server 20 or data center. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server 20, a data center, etc. that includes multiple available media integrated. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a digital versatile disc (DVD)), or a semiconductor medium (eg, a solid state disk (SSD)), etc.

[0073] When implemented by hardware or firmware, the aforementioned method flow is programmed into the hardware circuit to obtain the corresponding hardware circuit structure and realize the corresponding function. For example, a programmable logic device (PLD) (such as a field programmable gate array (FPGA)) is such an integrated circuit, and its logic function is determined by the user programming the device. The designer programs by himself to "integrate" a digital system on a PLD, without asking a chip manufacturer to design and make a dedicated integrated circuit chip. Moreover, nowadays, instead of manually making integrated circuit chips, this programming is mostly implemented by "logic compiler" software, which is similar to the software compiler used when writing program development, and the original code before compilation must also be written in a specific programming language, which is called hardware description language (HDL), and HDL is not just one, but many. Those skilled in the art should also be aware that it is only necessary to program the method flow slightly in the above-mentioned hardware description languages ​​and program it into the integrated circuit to easily obtain the hardware circuit that implements the logic method flow.

[0074] The embodiments described above are merely preferred embodiments of this specification and are not intended to limit the scope of this specification. Without departing from the design spirit of this specification, various modifications and improvements made to the technical solutions of this specification by ordinary technicians in this field should fall within the scope of protection determined by the claims of this specification.

Claims

1. A method for constructing a transcript knowledge graph, characterized in that: Includes steps: Read the transcript, extract several keywords from the transcript, and match the knowledge structure template according to the keywords; the knowledge structure template records the fields contained in the knowledge, Generate a number of pieces of knowledge according to the transcript and the knowledge structure template; According to the events involved in the knowledge, try to match the evidence on file, and if there is matching evidence on file, add the information of the evidence on file to the knowledge; According to the time and place involved in the knowledge, try to obtain the corresponding public area video surveillance clips; If there is a corresponding public area video surveillance clip, identifying an event in the public area video surveillance clip, and recording it as an identified event; Comparing the identified event with the event involved in the knowledge to obtain a matching degree, and adding the public area video surveillance clip and the matching degree to the knowledge; All acquired knowledge is represented using a graph to obtain a knowledge graph of the transcript.

2. A method for constructing a transcript knowledge graph according to claim 1, characterized in that: The method of extracting several keywords from the transcript includes: According to a number of preset first-class reference words, skip the first several paragraphs of the transcript; Continuously reading the transcript by paragraphs and extracting paragraph keywords of the paragraphs; Calculating the relevance between the subsequent paragraph keyword and the previous paragraph keyword obtained, and taking the maximum value of the relevance as the final value of the relevance; The paragraph keywords whose final relevance values ​​are less than a preset threshold are deleted, and the keywords of the transcript are obtained based on the remaining paragraph keywords.

3. A method for constructing a transcript knowledge graph according to claim 2, characterized in that: The method for matching the knowledge structure template according to the keyword includes: Reading a preset knowledge structure template, wherein the knowledge structure template is associated with a plurality of matching words; The matching degree between the keyword and the matching word is calculated, and a matching knowledge structure template is obtained according to the matching degree.

4. A method for constructing a transcript knowledge graph according to claim 3, characterized in that: Methods for presetting knowledge structure templates include: Receive manually set key information sheets and sample transcripts for various types of cases; Generate a knowledge structure template according to the key information table; Based on the sample transcript of the type of the case, a number of matching words are extracted.

5. A method for constructing a transcript knowledge graph according to any one of claims 1 to 4, characterized in that: Methods for matching evidence on record include: Reading a pre-configured reference evidence table, wherein the reference evidence table records several evidence types and evidence descriptions corresponding to the event; Obtaining several types of evidence and evidence descriptions corresponding to the event according to the reference evidence table; The evidence on file is compared with the evidence type. If the types match, the knowledge is matched with the evidence description. If they match, it is determined that the evidence on file matches.

6. A method for constructing a transcript knowledge graph according to any one of claims 1 to 4, characterized in that: The method of obtaining the corresponding public area video surveillance clip includes: obtaining relevant public areas of the places to which said knowledge relates; Searching for a corresponding surveillance camera according to the relevant area; If there is a corresponding surveillance camera, obtaining a video surveillance clip of the public area according to the time involved in the knowledge; in, The method of obtaining the relevant public area of ​​the place involved in the knowledge includes: Obtaining the location coordinates of the location involved in the knowledge; Reading pre-configured map data to obtain a public area near the location; It is determined whether the nearby public area can reach the location, and the public area that can be reached is used as the relevant public area of ​​the location involved in the knowledge.

7. A transcript knowledge graph construction system, characterized in that: include: A reading module, reading the transcript, extracting a number of keywords from the transcript, and matching the knowledge structure template according to the keywords; The knowledge structure template records the fields contained in the knowledge. A knowledge module generates a plurality of pieces of knowledge according to the transcript and the knowledge structure template; A first matching module attempts to match the on-record evidence according to the event involved in the knowledge, and if there is matching on-record evidence, adds the information of the on-record evidence to the knowledge; The second matching module attempts to obtain corresponding public area video surveillance clips according to the time and location involved in the knowledge; The recognition module recognizes an event in the public area video surveillance segment if there is a corresponding public area video surveillance segment, and records it as a recognition event; A matching module, which compares the identified event with the event involved in the knowledge, obtains a matching degree, and adds the public area video surveillance clip and the matching degree to the knowledge; The construction module represents all the acquired knowledge using a graph to obtain the transcript knowledge graph.

8. An electronic device, characterized in that including a processor and a memory; The processor is connected to the memory; The memory is used to store executable program code; The processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to execute the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, which, when executed by a processor, implements the method according to any one of claims 1 to 6.

Images