CN113590737A - Event data processing method, device, equipment and medium based on knowledge graph - Google Patents
Event data processing method, device, equipment and medium based on knowledge graph Download PDFInfo
- Publication number
- CN113590737A CN113590737A CN202111144445.3A CN202111144445A CN113590737A CN 113590737 A CN113590737 A CN 113590737A CN 202111144445 A CN202111144445 A CN 202111144445A CN 113590737 A CN113590737 A CN 113590737A
- Authority
- CN
- China
- Prior art keywords
- event
- entity
- linked list
- knowledge graph
- name
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/31—Indexing; Data structures therefor; Storage structures
- G06F16/316—Indexing structures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Computational Linguistics (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The application relates to a method, a device, equipment and a medium for processing event data based on a knowledge graph, wherein the method comprises the following steps: acquiring a query condition; utilizing a term pointer array of an event information knowledge graph called by query condition filtering to position a target array element corresponding to a query condition in the term pointer array; and matching each element in the event identifier list corresponding to the target linked list element by using the event pointer array of the event information knowledge graph to obtain a target second linked list corresponding to the matched event identifier so as to indicate the element information of the target event information corresponding to the event to be retrieved in the event information knowledge graph. The event intelligence knowledge graph comprises an event entity, a time entity, a place entity, a person entity, a cause entity, a passing entity and a result entity, and the relationship type of the graph comprises a time description relationship, a place description relationship, a participation description relationship and a development description relationship. The analysis and processing of the event intelligence data can be reliably completed.
Description
Technical Field
The present application relates to the field of data processing technologies, and in particular, to a method, an apparatus, a device, and a medium for processing event data based on a knowledge graph.
Background
Organization, storage and retrieval of information are important in information analysis processing, and are preconditions for correct information analysis processing. The event information is an important information category, and can provide data support for information analysts to the development law of historical problems and the deep insight of the current focus problem, so that the organization and the processing of the event information are particularly important. Knowledge Graph (Knowledge Graph) is a modern theoretical technology which combines theories and methods of applying mathematics, graphics, information visualization technology, information science and other subjects with methods of metrology introduction analysis, co-occurrence analysis and the like, and utilizes a visualized Graph to vividly show core structures, development histories, frontier fields and an overall Knowledge framework of the subjects to achieve the purpose of multi-subject fusion. The knowledge map is called knowledge domain visualization or knowledge domain mapping map in the book intelligence world, is a series of different graphs for displaying the relationship between the knowledge development process and the structure, describes knowledge resources and carriers thereof by using visualization technology, and excavates, analyzes, constructs, draws and displays knowledge and the mutual relation between the knowledge resources and the carriers.
In modern information analysis processing, the traditional information analysis processing method mainly comprises the following steps: the method comprises an intelligence correlation analysis method based on the knowledge graph, a network security emergency response method based on the knowledge graph, a comprehensive cognition method based on key figure target identification, a method for constructing the knowledge graph oriented to the intelligence analysis, a method for constructing the threat intelligence knowledge graph oriented to the text data and the like. However, in the process of implementing the present invention, the inventor finds that the conventional method for analyzing and processing the information has a technical problem that the analysis and processing of the event information data cannot be reliably completed.
Disclosure of Invention
In view of the above, it is necessary to provide a method for processing event data based on a knowledge graph, an apparatus for processing event data based on a knowledge graph, a computer device and a computer readable storage medium, which can reliably complete the analysis processing of event intelligence data.
In order to achieve the above purpose, the embodiment of the invention adopts the following technical scheme:
in one aspect, an embodiment of the present invention provides an event data processing method based on a knowledge graph, including:
acquiring query conditions and calling the constructed event information knowledge graph; the query condition is a keyword input aiming at an event to be retrieved, the entity type of the event intelligence knowledge graph comprises an event entity, a time entity, a place entity, a figure entity, a cause entity, a passing entity and a result entity, and the relationship type of the event intelligence knowledge graph comprises a time description relationship, a place description relationship, a participation description relationship and a development description relationship;
utilizing the query condition to filter a term pointer array of the event information knowledge graph, and positioning a target array element corresponding to the query condition in the term pointer array; the term pointer array is a pointer array formed by name terms of all entities of the event information knowledge graph;
searching a target linked list element of which the entity type is an event entity in the first linked list according to the first linked list corresponding to the pointer of the target array element; the first linked list is an external memory storage address for storing source information data for constructing an event information knowledge graph, an entity type of the event information knowledge graph and an internal memory storage linked list of an event identifier;
matching each element in an event identifier list corresponding to the target linked list element by using an event pointer array of an event information knowledge graph to obtain a target second linked list corresponding to the matched event identifier; the event pointer array is used for storing event identifications of all event entities of the event information knowledge graph, pointers of all elements of the event pointer array point to a second linked list respectively, each second linked list is used for storing element information of a group of event information, and the target second linked list is used for indicating element information of target event information corresponding to an event to be retrieved in the event information knowledge graph.
In another aspect, a knowledge-graph-based event data processing apparatus is also provided, including:
the retrieval input module is used for acquiring a query condition and calling the constructed event information knowledge graph; the query condition is a keyword input aiming at an event to be retrieved, the entity type of the event intelligence knowledge graph comprises an event entity, a time entity, a place entity, a figure entity, a cause entity, a passing entity and a result entity, and the relationship type of the event intelligence knowledge graph comprises a time description relationship, a place description relationship, a participation description relationship and a development description relationship;
the element positioning module is used for filtering a term pointer array of the event information knowledge graph by using the query condition and positioning a target array element corresponding to the query condition in the term pointer array; the term pointer array is a pointer array formed by name terms of all entities of the event information knowledge graph;
the linked list searching module is used for searching a target linked list element of which the entity type is an event entity in the first linked list according to the first linked list corresponding to the pointer of the target array element; the first linked list is an external memory storage address for storing source information data for constructing an event information knowledge graph, an entity type of the event information knowledge graph and an internal memory storage linked list of an event identifier;
the matching output module is used for matching each element in the event identifier list corresponding to the target linked list element by using the event pointer array of the event information knowledge graph to obtain a target second linked list corresponding to the matched event identifier; the event pointer array is used for storing event identifications of all event entities of the event information knowledge graph, pointers of all elements of the event pointer array point to a second linked list respectively, each second linked list is used for storing element information of a group of event information, and the target second linked list is used for indicating element information of target event information corresponding to an event to be retrieved in the event information knowledge graph.
In yet another aspect, a computer device is provided, which includes a memory and a processor, the memory stores a computer program, and the processor implements the steps of any one of the above-mentioned knowledge-graph-based event data processing methods when executing the computer program.
In yet another aspect, a computer readable storage medium is provided, on which a computer program is stored, which when executed by a processor implements the steps of any of the above-mentioned knowledge-graph based event data processing methods.
One of the above technical solutions has the following advantages and beneficial effects:
the event data processing method, the device, the equipment and the medium based on the knowledge graph utilize the event intelligence knowledge graph, which fully covers event elements such as time, place, people, cause, passing and result of an event, and correspondingly defines entity types and relationship types between entities in the knowledge graph. Compared with the traditional information knowledge graph, the event information knowledge graph is more suitable for accurately describing event information. The storage and index mode of the event information knowledge graph fully utilizes the design concepts of external memory storage information files and internal memory storage index information, so that the event information can be completely stored, the practical quick index capability can be provided, and the basic technical support is provided for efficiently retrieving the event information. By obtaining the given query terms and utilizing the event information knowledge graph for retrieval, the entity corresponding to the query terms in the event information knowledge graph can be quickly and accurately positioned, and all information related element information of the queried event is returned.
Drawings
FIG. 1 is a schematic flow diagram of a method for knowledge-graph based event data processing in one embodiment;
FIG. 2 is a schematic flow chart diagram of a method for knowledge-graph based event data processing in another embodiment;
FIG. 3 is a schematic diagram of an embodiment of a training process of an event intelligence ontology model;
FIG. 4 is a flow diagram illustrating the construction of an event intelligence knowledgegraph in one embodiment;
FIG. 5 is a flow diagram illustrating an embodiment of event intelligence knowledgegraph storage process;
FIG. 6 is a flow diagram illustrating an example of an event intelligence knowledgegraph indexing process;
FIG. 7 is a block diagram of an event data processing apparatus based on a knowledge-graph according to an embodiment.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein in the description of the present application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.
In addition, the technical solutions in the embodiments of the present invention may be combined with each other, but it must be based on the realization of those skilled in the art, and when the technical solutions are contradictory or cannot be realized, such a combination of technical solutions should be considered to be absent and not within the protection scope of the present invention.
In the traditional information analysis processing method, the information correlation analysis method based on the knowledge graph mainly comprises the steps of analyzing a downloaded information data TXT document, constructing a triple information knowledge base, and utilizing a SPARQL to inquire a return result. The network security emergency response method based on the knowledge graph is characterized in that a security information base and a knowledge graph base are constructed to match network security events, and an emergency processing scheme is provided. The comprehensive cognition method based on key character target recognition performs correlation analysis and mutual evidences on social attributes, activity rules and behavior habits of key characters by combining image character recognition and a constructed target knowledge graph. The method for constructing the knowledge graph for information analysis is to construct the knowledge graph for the collected data by utilizing text processing technologies such as data cleaning, entity identification and the like. The method for constructing the threat intelligence knowledge-graph oriented to the text data is characterized in that a triple is automatically extracted from the text threat intelligence data by aiming at threat information contained in a text and combining predefinition of threat types and threat relation types, and the triple is stored through a graph database so as to construct the threat intelligence knowledge-graph.
In view of the conventional knowledge mapping analysis technology, the inventor found that: (1) the existing knowledge graph-based information analysis method cannot be directly applied to the analysis of event information because the attribute elements of the event information are unique and generally need to contain elements such as time, place and people, while the traditional method can only meet the analysis processing of 1 to 2 event information elements and cannot fully support the required analysis capability. (2) The identification of event information is complex, and the expected effect cannot be achieved by using the traditional knowledge graph construction method. Most of the conventional methods adopt an entity identification method to construct a knowledge graph triple, and event intelligence has numerous elements and can be mistakenly constructed only by the entity identification, so that a more effective and reliable method is required to be utilized for processing.
In summary, the invention provides an event data processing method based on a knowledge graph aiming at the technical problem that the analysis processing of event information data can not be reliably completed in the traditional information analysis processing method, aiming at the event information data to be processed, the event information data is represented, stored and retrieved by using the knowledge graph technology, 6 elements of time, place, person, cause, pass, result and the like of the event are fully considered, an event information body is designed, the extracted entity is constructed into the event knowledge graph by using a machine learning method, and then the storage and retrieval design of the graph is further carried out based on the elements, so that the problem of reliable analysis processing of the event information data is effectively solved.
Referring to fig. 1, in one aspect, the present invention provides a method for processing event data based on a knowledge graph, including the following steps S12 to S18:
s12, obtaining query conditions and calling the constructed event intelligence knowledge graph; the query condition is a keyword input aiming at an event to be retrieved, the entity types of the event intelligence knowledge graph comprise an event entity, a time entity, a place entity, a figure entity, a cause entity, a passing entity and a result entity, and the relationship types of the event intelligence knowledge graph comprise a time description relationship, a place description relationship, a participation description relationship and a development description relationship.
It will be appreciated that the event intelligence knowledgegraph can be automatically constructed using given source intelligence data, i.e., given intelligence raw data (which is textual data). The event information knowledge graph is a knowledge graph constructed under an event information ontology model, and for convenience of understanding, for example, but not limited to, the event information knowledge graph can be recorded aseikgNamely, Event Intelligent Knowledge Graph.eikgThere are several types of entities: event entityen_eventFor expressing an event name; time entityen_timeTime attributes used to describe the event entities; venue entityen_locA location attribute for describing an event entity; persona entityen_figA person attribute for describing the event entity; causative entityen_causeA cause attribute for describing the event entity; passing through an entityen_courseA pass-through attribute for describing the event entity; fruit bodyen_retDescribing the result attributes of the event entity.
eikgThere are several relationship types: temporal descriptive relationshiprel_timeExpressing a relationship for describing a certain object by time; location description relationshipsrel_locA relationship for expressing that a certain object is described by a place; participate in describing relationshipsrel_partExpressing the relationship of people participating in a certain event; developing descriptive relationshipsrel_devpFor expressing the cause, passage, context development between the results of an event and evolution of an event into another eventAnd (4) unfolding relation.
S14, filtering the term pointer array of the event information knowledge graph by using the query condition, and positioning the target array element corresponding to the query condition in the term pointer array; the term pointer array is a pointer array formed by name terms of all entities of the event intelligence knowledge graph.
S16, searching the target linked list element of which the entity type is the event entity in the first linked list according to the first linked list corresponding to the pointer of the target array element; the first linked list is an external memory storage address for storing source information data for constructing the event information knowledge graph, the entity type of the event information knowledge graph and the internal memory storage linked list of the event identifier.
S18, matching each element in the event identifier list corresponding to the target linked list element by using the event pointer array of the event information knowledge map to obtain a target second linked list corresponding to the matched event identifier; the event pointer array is used for storing event identifications of all event entities of the event information knowledge graph, pointers of all elements of the event pointer array point to a second linked list respectively, each second linked list is used for storing element information of a group of event information, and the target second linked list is used for indicating element information of target event information corresponding to an event to be retrieved in the event information knowledge graph.
It is understood that the element information may include 6 elements of time, place, person, cause, passage, and result of the event. The source intelligence data is stored in an external memory, and the second linked list is also stored in the internal memory. For each event entity, a unique event identifier is distributed to the event intelligence knowledge graph in a storage mode.
The event data processing method based on the knowledge graph fully covers event elements such as time, place, people, causes, passing and results of the event, and the relationship type between the entity type and the entity in the knowledge graph is correspondingly defined. Compared with the traditional information knowledge graph, the event information knowledge graph is more suitable for accurately describing event information. The storage and index mode of the event information knowledge graph fully utilizes the design concepts of external memory storage information files and internal memory storage index information, so that the event information can be completely stored, the practical quick index capability can be provided, and the basic technical support is provided for efficiently retrieving the event information. By obtaining the given query terms and utilizing the event information knowledge graph for retrieval, the entity corresponding to the query terms in the event information knowledge graph can be quickly and accurately positioned, and all information related element information of the queried event is returned.
Referring to fig. 2, in an embodiment, before step S12, the method for processing event data based on a knowledge graph may further include the following processing steps S09 to S11 related to the construction of an event intelligence knowledge graph:
s09, acquiring source information data;
s10, training the event element recognition model by adopting a machine learning method according to the source information data to obtain an event information ontology model; the event information ontology model comprises a time description model, a place description model, a character description model, a cause description model, a pass description model and a result description model;
s11, according to the source information data and the event information ontology model, constructing an event information knowledge map.
It can be appreciated that the present example presents an event intelligence ontology model for the characteristics of event intelligence. For ease of understanding, the event information ontology model can be uniformly written aseioEvent Intelligent Ontology, consisting ofetime,eloc,efig,ecause,ecourse,eret) The corresponding meanings of the components are (time, place, person, cause, passage, result), respectively, wherein,etimeis time type data;elocthe event name string is a geographical name string or a geographical name string array and is used for describing that an event possibly exists in a group of places;efigthe character string is a character name character string or a character name character string array and is used for describing that a group of characters possibly exist in an event;ecausea set of words separated by semicolons, used to describe the cause of the event;ecoursea set of words separated by semicolons, used to describe the passage of events;ereta set of words separated by semicolons is used to describe the outcome of an event.
The machine learning method can be various machine learning methods suitable for training an event information ontology model in the field, and can be selected according to the requirements of training efficiency, calculated amount, precision and the like. The source intelligence data may be, but is not limited to being, obtained by online collection by a computing device, downloaded from a database, or manually pre-collected.
Through the steps, after the time, place, person and other element identification models are trained aiming at the original information text, the event information knowledge graph can be automatically constructed based on the models.
Referring to fig. 3, in an embodiment, the step S10 may specifically include the following processing steps S101 to S107:
s101, extracting target words from source information data to obtain word lists corresponding to the target words respectively; each word list comprises a keyword list, a time word list, a place name list, a person name list and a verb list;
s103, according to the source information data, using a domain expert to describe and label each word list, and marking out each element word which is described and takes the key words in the key word list as events; each element word comprises a time word, a place word, a figure word, a cause word, a passing word and a result word;
s105, vectorizing each word list and each element word by using a BERT model;
and S107, performing event element recognition model training by using an autoregressive model according to the source information data, the vectorized word lists and the element words to obtain an event information ontology model.
It is understood that a domain expert refers to a business expert in the field, and may provide answers and comments of terms and the like for a computer in the face of certain problematic intelligence terms, specialized terms and the like. The BERT model is an existing language model in the field of machine learning, and may be used to process word vectorization, in this embodiment, the BERT model is directly applied to perform required vectorization processing, and the processing procedure may be understood in the same manner with reference to the processing flow of the BERT model itself.
Specifically, for the source intelligence data, for easy understanding, the source intelligence data can be recorded as intelligence text dataIThe corresponding features are labeled similarly below. It should be noted that, as the specific english notation used for each feature is used as its code, it is not the only notation and limitation for the feature that one skilled in the art would understand, and other notation may be used as required by the description, and the following description of the embodiments is similar to the description of the feature. The above processing steps can be described in the following processing procedures:
1: to pairIExtracting the keywords to form a keyword listkwlist；
2: to pairIExtracting time words to form a time word listtwlist；
3: to pairIExtracting place name entities to form a place name word listlwlist；
4: to pairIExtracting name entities to form name word listfwlist；
5: to pairIExtracting verbs to form a verb listawlist；
6: labeling by domain experts, combiningIGiven iskwlistAndtwlistthe description is marked withkwlistTime words for events;
7: labeling by domain experts, combiningIGiven iskwlistAnd wlan, the description is marked outkwlistA place word for an event;
8: labeling by domain experts, combiningIGiven iskwlistAndfwlistthe description is marked withkwlistA character word for the event;
9: by domain expertsLabeling, combiningIGiven iskwlistAndawlistthe description is marked withkwlistWords that are the cause of the event;
10: labeling by domain experts, combiningIGiven iskwlistAndawlistthe description is marked withkwlistIs the term of the event;
11: labeling by domain experts, combiningIGiven iskwlistAndawlistthe description is marked withkwlistA result word for an event;
12: using BERT modelkwlist、twlist、lwlist、fwlistAndawlistthe marked time words, the marked place words, the marked figure words, the marked cause words, the marked words and the marked result words are all vectorized so as to facilitate the processing of the subsequent steps;
13: training time description models using autoregressive modelsm_timeI.e. givenI、kwlistAndtwlisttime description words can be mapped; the autoregressive model is an existing model in the field, and the processing process of the autoregressive model can be understood by referring to the existing processing flow of the autoregressive model;
14: training a location description model using an autoregressive modelm_locI.e. givenI、kwlistAndlwlistlocation description words may be mapped;
15: training character description model using autoregressive modelm_figI.e. givenI、kwlistAndfwlistthe character description words can be mapped;
16: training cause description model using autoregressive modelm_causeI.e. givenI、kwlistAndawlistcause description words may be mapped;
17: training a described model using an autoregressive modelm_courseI.e. givenI、kwlistAndawlistdescriptive words may be mapped;
18: describing a model using autoregressive model training resultsm_retI.e. givenI、kwlistAndawlistthe result description can be mapped outWords.
Through the steps, the quick training of the event information ontology model for the original information text is realized. The ontology model training method can train the recognition models (namely, the description models) for the time, the place, the characters, the causes, the processes and the results of the events by utilizing entity extraction and machine learning, so that the subsequently constructed event information knowledge graph is closer to the knowledge graph for analyzing the events compared with the traditional information knowledge graph, and the constructed event information knowledge graph is more accurate.
Referring to fig. 4, in an embodiment, the step S11 may specifically include the following processing steps S111 to S116:
s111, identifying a time descriptor of the event entity by using a time description model, taking the time descriptor as a name of the constructed time entity, and constructing a time description relation to connect the event entity and the time entity;
s112, identifying the location descriptor of the event entity by using the location description model, taking the location descriptor as the name of the constructed location entity, and constructing a location description relation to connect the event entity and the location entity;
s113, identifying a character descriptor of the event entity by using a character description model, taking the character descriptor as the name of the constructed character entity, and constructing a connection participating description relation between the event entity and the character entity;
s114, identifying a cause descriptor of the event entity by using the cause description model, using the cause descriptor as the name of the constructed cause entity, and constructing a development description relation to connect the event entity and the cause entity;
s115, recognizing a passing descriptor of the event entity by using the passing descriptor, and constructing a development description relationship to connect the event entity and the passing entity by using the passing descriptor as a name of the constructed passing entity;
and S116, identifying a result descriptor of the event entity by using the result description model, taking the result descriptor as the name of the constructed result entity, and constructing a development description relationship to connect the event entity and the result entity.
It will be appreciated that, as previously describedThe embodiment shows that for given informative text dataI：
1: to pairIExtracting the keywords to form a keyword listkwlist；
2: to pairIExtracting time words to form a time word listtwlist；
3: to pairIExtracting place name entities to form a place name word listlwlist；
4: to pairIExtracting name entities to form name word listfwlist；
5: to pairIExtracting verbs to form a verb listawlist；
6: construction of event information knowledge grapheikgThe steps can be specifically described as the following processing procedures:
7: building event entitiesen_eventEntity namekwlistThe combination of Chinese words; such askwlistWhen words "today", "car ride", "park" are included, the combination means: "today-car-park", i.e. the concatenation of words.
8: building time entitiesen_timeBy usingm_timeIdentifying event entitiesen_eventAs a time descriptoren_timeName of entity, and constructrel_timeConnection ofen_eventAnden_time(ii) a Wherein the content of the first and second substances,m_timecan be considered a time recognition method, as in the above "today", usingm_time"today" can be identified as a specific date such as year, month, day, etc. Whilerel_timeFor describing the relationship between the year, month and day of the above-mentioned conversion and the event itself, e.g.rel_timeIt can be expressed that the time is the "time of occurrence" of the event.
9: building site entitiesen_locBy usingm_locIdentifying event entitiesen_eventAs a location descriptoren_locName of entity, and constructrel_locConnection ofen_eventAnden_loc；
10: building human entitiesen_figBy usingm_figIdentifying event entitiesen_eventAs a character descriptoren_figName of entity, and constructrel_partConnection ofen_eventAnden_fig；
11: construction of causative entitiesen_causeBy usingm_causeIdentifying event entitiesen_eventAs a cause descriptor ofen_causeName of entity, and constructrel_devpConnection ofen_eventAnden_cause；
12: building a Via entityen_courseBy usingm_courseIdentifying event entitiesen_eventAs a descriptor ofen_courseName of entity, and constructrel_devpConnection ofen_eventAnden_course；
13: construction of fruit entitiesen_retBy usingm_retIdentifying event entitiesen_eventThe result descriptor asen_retName of entity, and constructrel_devpConnection ofen_eventAnden_ret。
through the steps, the purpose of automatically constructing the event information knowledge graph based on the event information ontology model is achieved. The construction mode utilizes the ontology model extracted by the entity and trained before to directly generate the event information knowledge graph for the given information text, is a relatively intelligent construction means, and compared with the traditional information knowledge graph construction method, the method has the advantages that the event elements are more accurately generated by utilizing the machine learning model, and the construction speed is higher than that of the traditional construction method.
Referring to fig. 5, in an embodiment, before step S12, the method for processing event data based on a knowledge graph may further include the following processing steps S21 to S27:
s21, storing the source information data to the set external memory address;
s22, extracting the name words of all entities in the event information knowledge graph to form a name word list;
s23, forming a word pointer array by using all words in the name word list; pointers of the word pointer array point to the first linked list;
s24, setting the 1 st element of the first linked list as an external memory storage address;
s25, for each name word corresponding to the first linked list, respectively finding each name word in the event intelligence knowledge graph as all entities with entity names to form a name entity list corresponding to each name word;
s26, generating an element structure in the first linked list according to each entity in the name entity list; from the 2 nd element in the first linked list, each element structure is the entity type and the event identifier of the entity >;
and S27, taking out all event entities in the event intelligence knowledge graph, endowing each event entity with a unique event identifier, and inserting the event identifier of each event entity into the first linked list.
Specifically, the above processing steps can be described as the following processing procedures:
1: text data of intelligenceIPerforming external memory with a memory address ofaddr_I；
2: will be provided witheikgTaking out the name words of all the entities to form a name word listlist_w；
3：list_wAll the words in (1)wForm a word pointer arrayary_wWhose array element is thatlist_wWord in (1)wThe pointers of the array point to a first linked listlik(memory storage);
4: linked listlikWherein the 1 st element isaddr_I；
5: linked listlikStarting from the 2 nd element, each element is structured as<type_en,eventid>；
6: for each wordwCorresponding tolikTo find out the wordwAll entities appearing as entity names, forming eachwRespective corresponding entity listslist_w_en；
7: to is directed atlist_w_enEach entity in (1)enGenerating a linked listlikStructure of element (1)<type_en,eventid>Whereintype_enIs composed ofenThe type of the corresponding entity is set as,eventidis composed ofenCorresponding event identification (see next step 8), and inserting the linked list after the elements are generatedlikPerforming the following steps;
8: will be provided witheikgAll event type entities are taken out and assignedEach event type entity is given a unique identifier aseventid。
Through the steps, the storage processing of the event information knowledge graph is realized.
Referring to fig. 6, in an embodiment, before step S12, the method for processing event data based on a knowledge graph may further include the following processing steps S31 to S39:
s31, forming an event pointer array by using all event entities in the event information knowledge graph; pointers of each element of the event pointer array point to each second linked list respectively; the second linked list is a memory storage linked list;
s32, setting the 1 st element of each second linked list as the name of the event entity to which the element belongs;
s33, setting the 2 nd element of each second linked list as the name of the time entity corresponding to the event entity;
s34, setting the 3 rd element of each second linked list as the name of the place entity corresponding to the event entity;
s35, setting the 4 th element of each second linked list as the name of the person entity corresponding to the event entity to which the element belongs;
s36, setting the 5 th element of each second linked list as the name of the cause entity corresponding to the event entity;
s37, setting the 6 th element of each second linked list as the name of the passing entity corresponding to the event entity to which the element belongs;
s38, setting the 7 th element of each second linked list as the name of the result entity corresponding to the event entity;
and S39, setting the 8 th element of each second linked list as an external memory storage address.
Specifically, the above processing steps can be described as the following processing procedures:
1: all event type entities are formed into an event pointer arrayary_enFor each array element, the element content is the unique identifier of the event entity corresponding to the elementeventidThe element's pointer pointing to a second linked listglik(memory storage); it will be appreciated that if there are more than one element, then the second linked listglikCorrespond toThere are a plurality of, the same way of setting for each and different contents;
2：glikthe 1 st element is the name of the event entity corresponding to the linked list (namely the affiliated event entity);
3：glikthe 2 nd element is the name of the time entity corresponding to the event entity;
4：glikthe 3 rd element is the name of the place entity corresponding to the event entity;
5：glikthe 4 th element is the name of the person entity corresponding to the event entity to which the element belongs;
6：glikthe 5 th element is the name of the cause entity corresponding to the event entity;
7：glikthe 6 th element is the name of a passing entity corresponding to the event entity to which the element belongs;
8：glikthe 7 th element is the name of the result entity corresponding to the event entity;
9：glikthe 8 th element isaddr_I。
Through the steps, the index setting processing of the event information knowledge graph is realized.
In one embodiment, with respect to the above-described knowledge-graph-based event data processing method, the following retrieval process example may be given to make it easier to understand the contents of the above-described method:
for query conditionsqwI.e. querying the graph of the event according to a certain keyword:
1: by usingqwFiltering array of word pointersary_wLocate the corresponding array elementq；
2: along the elementqLinked list corresponding to the pointer oflikLooking up entity typestype_enAs event entitiesen_eventThe target linked list element of (2) sets the corresponding event identifier foundeventidIs listed aslist_eventid；
3: to is directed atlist_eventidEach element ofeventidUsing event pointer arrayary_enMatching to obtain corresponding linked listglikAnd therefore, the finding of the event knowledge graph is completed.
It should be understood that although the steps in the flowcharts of fig. 1-6 are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least some of the steps of fig. 1-6 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, and the order of performing the sub-steps or stages is not necessarily sequential, but may be performed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.
Referring to fig. 7, in one embodiment, there is further provided a knowledge-graph-based event data processing apparatus 100, which includes a retrieval input module 11, an element location module 13, a linked list lookup module 15, and a matching output module 17. The retrieval input module 11 is used for acquiring a query condition and calling a constructed event information knowledge graph; the query condition is a keyword input aiming at an event to be retrieved, the entity types of the event intelligence knowledge graph comprise an event entity, a time entity, a place entity, a figure entity, a cause entity, a passing entity and a result entity, and the relationship types of the event intelligence knowledge graph comprise a time description relationship, a place description relationship, a participation description relationship and a development description relationship. The element positioning module 13 is configured to filter a term pointer array of the event information knowledge graph by using the query condition, and position a target array element corresponding to the query condition in the term pointer array; the term pointer array is a pointer array formed by name terms of all entities of the event intelligence knowledge graph.
The linked list searching module 15 is configured to search, according to a first linked list corresponding to a pointer of a target array element, a target linked list element in which an entity type in the first linked list is an event entity; the first linked list is an external memory storage address for storing source information data for constructing the event information knowledge graph, the entity type of the event information knowledge graph and the internal memory storage linked list of the event identifier. The matching output module 17 is configured to match each element in the event identifier list corresponding to the target linked list element by using an event pointer array of the event information knowledge graph, so as to obtain a target second linked list corresponding to the matched event identifier; the event pointer array is used for storing event identifications of all event entities of the event information knowledge graph, pointers of all elements of the event pointer array point to a second linked list respectively, each second linked list is used for storing element information of a group of event information, and the target second linked list is used for indicating element information of target event information corresponding to an event to be retrieved in the event information knowledge graph.
The event data processing apparatus 100 based on the knowledgegraph uses the event information knowledgegraph, which sufficiently covers the event elements such as the time, the place, the person, the cause, the passage, the result, and the like of the event through the cooperation of the modules, and accordingly defines the entity type in the knowledgegraph and the relationship type between the entities. Compared with the traditional information knowledge graph, the event information knowledge graph is more suitable for accurately describing event information. The storage and index mode of the event information knowledge graph fully utilizes the design concepts of external memory storage information files and internal memory storage index information, so that the event information can be completely stored, the practical quick index capability can be provided, and the basic technical support is provided for efficiently retrieving the event information. By obtaining the given query terms and utilizing the event information knowledge graph for retrieval, the entity corresponding to the query terms in the event information knowledge graph can be quickly and accurately positioned, and all information related element information of the queried event is returned.
In one embodiment, the entity candidate module 17 may include a data acquisition module, an ontology training module, and a graph construction module. The data acquisition module is used for acquiring source intelligence data. The ontology training module is used for performing event element recognition model training by adopting a machine learning method according to source information data to obtain an event information ontology model; the event intelligence ontology model comprises a time description model, a place description model, a person description model, a cause description model, a pass description model and a result description model. The map building module is used for building an event information knowledge map according to the source information data and the event information ontology model.
In one embodiment, the modules of the entity candidate module 17 may be further configured to implement other corresponding sub-steps in the embodiments of the event data processing method based on the knowledge-graph.
In one embodiment, the event data processing apparatus 100 may further include other modules, which are used to implement other steps added in the embodiments of the event data processing method based on the knowledge graph.
For specific limitations of the event data processing apparatus 100 based on the knowledge graph, reference may be made to the corresponding limitations of the event data processing method based on the knowledge graph, and details thereof are not repeated here. The various modules in the above-described knowledge-graph based event data processing apparatus 100 may be implemented in whole or in part by software, hardware, and combinations thereof. The modules may be embedded in a hardware form or a device independent of a specific data processing function, or may be stored in a memory of the device in a software form, so that a processor may invoke and execute operations corresponding to the modules, where the device may be, but is not limited to, various data calculation and analysis devices existing in the art.
In still another aspect, a computer device is provided, which includes a memory and a processor, the memory stores a computer program, and the processor executes the computer program to implement the following steps: acquiring query conditions and calling the constructed event information knowledge graph; the query condition is a keyword input aiming at an event to be retrieved, the entity type of the event information knowledge graph comprises an event entity, a time entity, a place entity, a person entity, a cause entity, a passing entity and a result entity, and the relationship type of the event information knowledge graph comprises a time description relationship, a place description relationship, a participation description relationship and a development description relationship;
utilizing the query condition to filter a term pointer array of the event information knowledge graph, and positioning a target array element corresponding to the query condition in the term pointer array; the term pointer array is a pointer array formed by name terms of all entities of the event information knowledge graph; searching a target linked list element of which the entity type is an event entity in the first linked list according to the first linked list corresponding to the pointer of the target array element; the first linked list is an external memory storage address for storing source information data for constructing an event information knowledge graph, an entity type of the event information knowledge graph and an internal memory storage linked list of an event identifier;
matching each element in an event identifier list corresponding to the target linked list element by using an event pointer array of an event information knowledge graph to obtain a target second linked list corresponding to the matched event identifier; the event pointer array is used for storing event identifications of all event entities of the event information knowledge graph, pointers of all elements of the event pointer array point to a second linked list respectively, each second linked list is used for storing element information of a group of event information, and the target second linked list is used for indicating element information of corresponding target event information of an event to be retrieved in the event information knowledge graph.
In one embodiment, the processor when executing the computer program may also implement the additional steps or sub-steps of the above-described method for processing event data based on a knowledge-graph.
In yet another aspect, there is also provided a computer readable storage medium having a computer program stored thereon, the computer program when executed by a processor implementing the steps of: acquiring query conditions and calling the constructed event information knowledge graph; the query condition is a keyword input aiming at an event to be retrieved, the entity type of the event information knowledge graph comprises an event entity, a time entity, a place entity, a person entity, a cause entity, a passing entity and a result entity, and the relationship type of the event information knowledge graph comprises a time description relationship, a place description relationship, a participation description relationship and a development description relationship;
utilizing the query condition to filter a term pointer array of the event information knowledge graph, and positioning a target array element corresponding to the query condition in the term pointer array; the term pointer array is a pointer array formed by name terms of all entities of the event information knowledge graph; searching a target linked list element of which the entity type is an event entity in the first linked list according to the first linked list corresponding to the pointer of the target array element; the first linked list is an external memory storage address for storing source information data for constructing an event information knowledge graph, an entity type of the event information knowledge graph and an internal memory storage linked list of an event identifier;
matching each element in an event identifier list corresponding to the target linked list element by using an event pointer array of an event information knowledge graph to obtain a target second linked list corresponding to the matched event identifier; the event pointer array is used for storing event identifications of all event entities of the event information knowledge graph, pointers of all elements of the event pointer array point to a second linked list respectively, each second linked list is used for storing element information of a group of event information, and the target second linked list is used for indicating element information of corresponding target event information of an event to be retrieved in the event information knowledge graph.
In one embodiment, the computer program, when executed by the processor, may further implement the additional steps or sub-steps of the above-described method for processing event data based on a knowledge-graph.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware related to instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), synchronous link DRAM (Synchlink) DRAM (SLDRAM), Rambus DRAM (RDRAM), and interface DRAM (DRDRAM).
The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.
The above examples only express several embodiments of the present application, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for those skilled in the art, various changes and modifications can be made without departing from the spirit of the present application, and all of them fall within the scope of the present application. Therefore, the protection scope of the present patent should be subject to the appended claims.
Claims (10)
1. A knowledge graph-based event data processing method is characterized by comprising the following steps:
acquiring query conditions and calling the constructed event information knowledge graph; the query condition is a keyword input aiming at an event to be retrieved, the entity type of the event intelligence knowledge graph comprises an event entity, a time entity, a place entity, a person entity, a cause entity, a passing entity and a result entity, and the relationship type of the event intelligence knowledge graph comprises a time description relationship, a place description relationship, a participation description relationship and a development description relationship;
filtering a term pointer array of the event information knowledge graph by using the query condition, and positioning a target array element corresponding to the query condition in the term pointer array; the term pointer array is a pointer array formed by name terms of all entities of the event information knowledge graph;
searching a target linked list element of which the entity type is an event entity in the first linked list according to the first linked list corresponding to the pointer of the target array element; the first linked list is an external memory storage address for storing source information data for constructing the event information knowledge graph, the entity type of the event information knowledge graph and an internal memory storage linked list of the event identifier;
matching each element in the event identifier list corresponding to the target linked list element by using the event pointer array of the event information knowledge graph to obtain a target second linked list corresponding to the matched event identifier; the event pointer array is used for storing event identifications of all event entities of the event intelligence knowledge graph, pointers of all elements of the event pointer array point to a second linked list respectively, each second linked list is used for storing element information of a group of event intelligence, and the target second linked list is used for indicating the element information of corresponding target event intelligence of the event to be retrieved in the event intelligence knowledge graph.
2. The method for event data processing based on a knowledge-graph of claim 1, wherein before the steps of obtaining query conditions and invoking the constructed event intelligence knowledge-graph, the method further comprises:
obtaining the source intelligence data;
according to the source information data, performing event element recognition model training by adopting a machine learning method to obtain an event information ontology model; the event intelligence ontology model comprises a time description model, a place description model, a character description model, a cause description model, a pass description model and a result description model;
and constructing the event intelligence knowledge graph according to the source intelligence data and the event intelligence ontology model.
3. The event data processing method based on the knowledge graph according to claim 2, wherein the step of performing event element recognition model training by using a machine learning method according to the source information data to obtain an event information ontology model comprises:
extracting target words from the source information data to obtain word lists corresponding to the target words respectively; each word list comprises a keyword list, a time word list, a place name list, a person name word list and a verb list;
according to the source information data, using a domain expert to describe and label each word list, and marking out each element word which describes an event by taking a keyword in the keyword list as an event; each element word comprises a time word, a place word, a figure word, a cause word, a passing word and a result word;
vectorizing each word list and each element word by using a BERT model;
and performing event element recognition model training by using an autoregressive model according to the source information data, each vectorized word list and each element word to obtain the event information ontology model.
4. The knowledgegraph-based event data processing method of claim 3, wherein the step of constructing the event intelligence knowledgegraph from the source intelligence data and the event intelligence ontology model comprises:
constructing the event entity by using the keyword list; the entity name of the event entity is a combination of keywords in the keyword list;
identifying a time descriptor of the event entity by using the time description model, taking the time descriptor as a name of the constructed time entity, and constructing the time description relation to connect the event entity and the time entity;
identifying a location descriptor of the event entity by using the location description model, taking the location descriptor as a name of the constructed location entity, and constructing the location description relation to connect the event entity and the location entity;
identifying a character descriptor of the event entity by using the character description model, taking the character descriptor as the name of the constructed character entity, and constructing the participation description relationship to connect the event entity and the character entity;
identifying cause descriptors of the event entities by using the cause description model, using the cause descriptors as names of the constructed cause entities, and constructing the development description relationship to connect the event entities and the cause entities;
identifying a via descriptor of the event entity by using the via description model, and constructing the development description relationship to connect the event entity and the via entity by using the via descriptor as a name of the constructed via entity;
and identifying a result descriptor of the event entity by using the result description model, taking the result descriptor as the name of the constructed result entity, and constructing the development description relation to connect the event entity and the result entity.
5. The knowledgegraph-based event data processing method according to any one of claims 1 to 4, wherein before the step of obtaining query conditions and invoking the constructed event intelligence knowledgegraph, the method further comprises:
storing the source information data to a set external memory storage address;
taking out name words of all entities in the event information knowledge graph to form a name word list;
forming a word pointer array by using all words in the name word list; pointers of the word pointer array point to the first linked list;
setting the 1 st element of the first linked list as the external memory storage address;
for each name word corresponding to the first linked list, respectively finding each name word in the event intelligence knowledge graph as all entities with entity names to form a name entity list corresponding to each name word;
generating an element structure in the first linked list according to each entity in the name entity list; from the 2 nd element in the first linked list, each element structure is an entity type and an event identifier of the entity;
and taking out all event entities in the event intelligence knowledge graph, endowing each event entity with a unique event identifier, and inserting the event identifier of each event entity into the first linked list.
6. The knowledgegraph-based event data processing method of claim 5, wherein before the steps of obtaining query conditions and invoking the constructed event intelligence knowledgegraph, further comprising:
forming an event pointer array by using all event entities in the event information knowledge graph; pointers of each element of the event pointer array point to each second linked list respectively; the second linked list is a memory storage linked list;
setting the 1 st element of each second linked list as the name of the event entity to which the element belongs;
setting the 2 nd element of each second linked list as the name of the time entity corresponding to the event entity to which the element belongs;
setting the 3 rd element of each second linked list as the name of the place entity corresponding to the event entity;
setting the 4 th element of each second linked list as the name of the person entity corresponding to the event entity to which the element belongs;
setting the 5 th element of each second linked list as the name of the cause entity corresponding to the event entity;
setting the 6 th element of each second linked list as the name of a passing entity corresponding to the event entity to which the element belongs;
setting the 7 th element of each second linked list as the name of a result entity corresponding to the event entity to which the element belongs;
and setting the 8 th element of each second linked list as the external memory storage address.
7. A knowledge-graph based event data processing apparatus, comprising:
the retrieval input module is used for acquiring a query condition and calling the constructed event information knowledge graph; the query condition is a keyword input aiming at an event to be retrieved, the entity type of the event intelligence knowledge graph comprises an event entity, a time entity, a place entity, a person entity, a cause entity, a passing entity and a result entity, and the relationship type of the event intelligence knowledge graph comprises a time description relationship, a place description relationship, a participation description relationship and a development description relationship;
the element positioning module is used for filtering a term pointer array of the event information knowledge graph by using the query condition and positioning a target array element corresponding to the query condition in the term pointer array; the term pointer array is a pointer array formed by name terms of all entities of the event information knowledge graph;
the linked list searching module is used for searching a target linked list element of which the entity type is an event entity in the first linked list according to the first linked list corresponding to the pointer of the target array element; the first linked list is an external memory storage address for storing source information data for constructing the event information knowledge graph, the entity type of the event information knowledge graph and an internal memory storage linked list of the event identifier;
the matching output module is used for matching each element in the event identifier list corresponding to the target linked list element by using the event pointer array of the event information knowledge graph to obtain a target second linked list corresponding to the matched event identifier; the event pointer array is used for storing event identifications of all event entities of the event intelligence knowledge graph, pointers of all elements of the event pointer array point to a second linked list respectively, each second linked list is used for storing element information of a group of event intelligence, and the target second linked list is used for indicating the element information of corresponding target event intelligence of the event to be retrieved in the event intelligence knowledge graph.
8. The knowledge-graph-based event data processing apparatus of claim 7, further comprising:
the data acquisition module is used for acquiring the source information data;
the ontology training module is used for performing event element recognition model training by adopting a machine learning method according to the source information data to obtain an event information ontology model; the event intelligence ontology model comprises a time description model, a place description model, a character description model, a cause description model, a pass description model and a result description model;
and the map building module is used for building the event information knowledge map according to the source information data and the event information ontology model.
9. A computer device comprising a memory and a processor, the memory storing a computer program, wherein the processor when executing the computer program implements the steps of the method of knowledge-graph based event data processing of any one of claims 1 to 6.
10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method for knowledge-graph based event data processing according to any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111144445.3A CN113590737B (en) | 2021-09-28 | 2021-09-28 | Event data processing method, device, equipment and medium based on knowledge graph |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111144445.3A CN113590737B (en) | 2021-09-28 | 2021-09-28 | Event data processing method, device, equipment and medium based on knowledge graph |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113590737A true CN113590737A (en) | 2021-11-02 |
| CN113590737B CN113590737B (en) | 2021-12-17 |
Family
ID=78242307
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111144445.3A Active CN113590737B (en) | 2021-09-28 | 2021-09-28 | Event data processing method, device, equipment and medium based on knowledge graph |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113590737B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114706992A (en) * | 2022-02-17 | 2022-07-05 | 中科雨辰科技有限公司 | Event information processing system based on knowledge graph |
| CN114969383A (en) * | 2022-08-02 | 2022-08-30 | 深圳易伙科技有限责任公司 | Application processing method and device based on zero code development |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104866593A (en) * | 2015-05-29 | 2015-08-26 | 中国电子科技集团公司第二十八研究所 | Database searching method based on knowledge graph |
| US10055450B1 (en) * | 2014-08-19 | 2018-08-21 | Abdullah Uz Tansel | Efficient management of temporal knowledge |
| CN110851616A (en) * | 2019-10-08 | 2020-02-28 | 杭州电子科技大学 | RDF knowledge graph storage and management method based on domain subgraphs |
| CN112948596A (en) * | 2021-04-01 | 2021-06-11 | 泰豪软件股份有限公司 | Knowledge graph construction method and device, computer equipment and computer storage medium |
-
2021
- 2021-09-28 CN CN202111144445.3A patent/CN113590737B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10055450B1 (en) * | 2014-08-19 | 2018-08-21 | Abdullah Uz Tansel | Efficient management of temporal knowledge |
| CN104866593A (en) * | 2015-05-29 | 2015-08-26 | 中国电子科技集团公司第二十八研究所 | Database searching method based on knowledge graph |
| CN110851616A (en) * | 2019-10-08 | 2020-02-28 | 杭州电子科技大学 | RDF knowledge graph storage and management method based on domain subgraphs |
| CN112948596A (en) * | 2021-04-01 | 2021-06-11 | 泰豪软件股份有限公司 | Knowledge graph construction method and device, computer equipment and computer storage medium |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114706992A (en) * | 2022-02-17 | 2022-07-05 | 中科雨辰科技有限公司 | Event information processing system based on knowledge graph |
| CN114706992B (en) * | 2022-02-17 | 2022-09-30 | 中科雨辰科技有限公司 | Event information processing system based on knowledge graph |
| CN114969383A (en) * | 2022-08-02 | 2022-08-30 | 深圳易伙科技有限责任公司 | Application processing method and device based on zero code development |
| CN114969383B (en) * | 2022-08-02 | 2022-10-25 | 深圳易伙科技有限责任公司 | Application processing method and device based on zero code development |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113590737B (en) | 2021-12-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110765257B (en) | Intelligent consulting system of law of knowledge map driving type | |
| CN111353310B (en) | Named entity identification method and device based on artificial intelligence and electronic equipment | |
| WO2021213314A1 (en) | Data processing method and device, and computer readable storage medium | |
| CN111026671B (en) | Test case set construction method and test method based on test case set | |
| CN113590737B (en) | Event data processing method, device, equipment and medium based on knowledge graph | |
| CN110968699A (en) | Logic map construction and early warning method and device based on event recommendation | |
| CN110457431A (en) | Answering method, device, computer equipment and the storage medium of knowledge based map | |
| US20120284259A1 (en) | Automated Generation of Ontologies | |
| Zirn et al. | Multidimensional topic analysis in political texts | |
| WO2020010834A1 (en) | Faq question and answer library generalization method, apparatus, and device | |
| US7478192B2 (en) | Network of networks of associative memory networks | |
| CN113806513A (en) | Question-answering system construction method and system based on knowledge graph in military field | |
| Zhang et al. | Foundations of intelligent knowledge management | |
| CN112966053B (en) | Knowledge graph-based marine field expert database construction method and device | |
| CN111696656B (en) | Doctor evaluation method and device of Internet medical platform | |
| CN113254651B (en) | Method and device for analyzing referee document, computer equipment and storage medium | |
| Wu et al. | PaintKG: the painting knowledge graph using bilstm-crf | |
| CN109766442A (en) | method and system for classifying user notes | |
| CN117312531A (en) | Power distribution network fault attribution analysis method based on large language model with enhanced knowledge graph | |
| CN107943937A (en) | A kind of debtors assets monitoring method and system based on trial open information analysis | |
| CN116701648A (en) | Mapping knowledge graph and schema design method based on standard specification | |
| CN115269806A (en) | Question-answering method, electronic device and storage medium applied to mineral domain knowledge graph | |
| CN115905677A (en) | Intelligent search system for medical field | |
| CN111401055B (en) | Method and apparatus for extracting context information from financial information | |
| CN107679154B (en) | Method, system and medium for solving historical problems based on time axis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |