CN113934764A - Event information processing method and device and electronic equipment - Google Patents
Event information processing method and device and electronic equipment Download PDFInfo
- Publication number
- CN113934764A CN113934764A CN202010603795.0A CN202010603795A CN113934764A CN 113934764 A CN113934764 A CN 113934764A CN 202010603795 A CN202010603795 A CN 202010603795A CN 113934764 A CN113934764 A CN 113934764A
- Authority
- CN
- China
- Prior art keywords
- event
- elements
- information
- incidence relation
- graph
- 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.)
- Pending
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
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2457—Query processing with adaptation to user needs
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Animal Behavior & Ethology (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention provides a method and a device for processing event information and electronic equipment, wherein the method comprises the following steps: extracting information of event description information of an event, and acquiring a plurality of event elements of the event and a first incidence relation among the event elements; giving an event identifier to the event, and establishing a second association relationship between all or part of event elements of the event and the event identifier; and constructing nodes in an event graph by using the event identification and the event elements of the event, constructing edges in the event graph by using the first incidence relation and the second incidence relation, and generating the event graph for the event. According to the embodiment of the invention, the event elements and the incidence relation among the elements are extracted from the event description information, the event map is constructed based on the event elements, and the event information is stored by adopting the structured topological structure, so that the subsequent inquiry and analysis related to the incidence relation are conveniently carried out on the event.
Description
Technical Field
The application relates to an event information processing method and device and electronic equipment, and belongs to the technical field of computers.
Background
With the development of information technology, a large number of events occur each day, and the events are generally stored in a server of a website or a database in a text form, and the events may relate to various fields, such as daily news events, various transaction events on an e-commerce platform, various alarm events of a cloud service platform, and the like. The recording of events is for the purpose of various queries and analyses of events at a later date.
In the prior art, various event information storage modes are scattered, and are generally stored directly in a text form. In terms of the requirement of retrieving and analyzing events, the events often need to be analyzed in a serial-parallel manner, for example, it is desirable to find events with similar characteristics, or events involving the same people or places. For the search and analysis requirements, in the prior art, the serial-parallel operation is often performed based on database search and manual comparison, the processing efficiency is not high, and the accuracy and the recall rate are difficult to ensure along with the increase of the event information quantity.
Disclosure of Invention
The embodiment of the invention provides an event information processing method and device and electronic equipment, and aims to improve the processing efficiency of event query.
In order to achieve the above object, an embodiment of the present invention provides an event information processing method, including:
extracting information of event description information of an event, and acquiring a plurality of event elements of the event and a first incidence relation among the event elements;
giving an event identifier to the event, and establishing a second association relationship between all or part of event elements of the event and the event identifier; and constructing nodes in an event graph by using the event identification and the event elements of the event, constructing edges in the event graph by using the first incidence relation and the second incidence relation, and generating the event graph for the event.
The embodiment of the invention also provides a method for processing the event information, which comprises the following steps:
acquiring a query condition of a user, wherein the query condition comprises event elements and/or a first incidence relation among the event elements;
traversing an event graph according to the query condition, and determining a first node and/or edge corresponding to the event element and/or the first incidence relation;
determining a second node which is associated with the first node and/or the first edge and corresponds to the event identifier according to the first node and/or the first edge, and acquiring the event identifier from the second node;
and acquiring corresponding event description information according to the event identifier, and returning the event description information to the user.
An embodiment of the present invention further provides an event information processing apparatus, including:
the information extraction module is used for extracting information of event description information of an event and acquiring a plurality of event elements of the event and a first incidence relation among the event elements;
the event identifier processing module is used for endowing the event with an event identifier and establishing a second association relationship between all or part of event elements of the event and the event identifier;
and the graph generation module is used for constructing nodes in the event graph by using the event identifications and the event elements of the events, constructing edges in the event graph by using the first incidence relation and the second incidence relation, and generating the event graph for the events.
An embodiment of the present invention further provides an event information processing apparatus, including:
the query condition acquisition module is used for acquiring a query condition of a user, wherein the query condition comprises event elements and/or a first incidence relation among the event elements;
the traversal query module is used for traversing the event graph according to the query condition and determining a first node and/or edge corresponding to the event element and/or the first incidence relation;
an event identifier determining module, configured to determine, according to the first node and/or the first edge, a second node associated with the first node and/or the first edge and corresponding to an event identifier, and obtain the event identifier from the second node;
and the query result feedback module is used for acquiring corresponding event description information according to the event identifier and returning the event description information to the user.
An embodiment of the present invention further provides an electronic device, including:
a memory for storing a program;
and the processor is used for operating the program stored in the memory so as to execute the processing method of the event information.
According to the event information processing method, the event information processing device and the electronic equipment, the event elements and the incidence relation among the elements are extracted from the event description information, the event map is constructed based on the event elements, and the event information is stored by adopting the structured topological structure, so that the subsequent inquiry and analysis related to the incidence relation are conveniently carried out on the event.
The foregoing description is only an overview of the technical solutions of the present invention, and the embodiments of the present invention are described below in order to make the technical means of the present invention more clearly understood and to make the above and other objects, features, and advantages of the present invention more clearly understandable.
Drawings
FIG. 1 is a flowchart illustrating a method for processing event information according to an embodiment of the present invention;
FIG. 2 is a schematic diagram of an event graph structure according to an embodiment of the present invention;
FIG. 3 is a second flowchart illustrating a method for processing event information according to an embodiment of the present invention;
FIG. 4 is a schematic structural diagram of an event information processing apparatus according to an embodiment of the present invention;
FIG. 5 is a second schematic structural diagram of an event information processing apparatus according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.
Detailed Description
Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.
The embodiment of the invention provides a method for processing event information, which extracts a plurality of event elements and incidence relations among the elements by extracting information of event description information, and generates an event graph according to the event elements and the incidence relations, wherein each event element corresponds to each node (called as an event element node) in the event graph, and the incidence relations among the event elements are embodied by edges connecting each node in the event graph. Each event is assigned with a unique event identifier, the event identifier also corresponds to a node (called an event identifier node) in the event graph, and the event identifier node is connected with an event element node of the event through an edge, so that the event content contained in a certain event is embodied.
The event graph is formed by elements of a plurality of events, in the event graph, the same elements correspond to the same element node, that is, the situation that the plurality of events contain the same element node exists, on one hand, the formed event graph records the content of key elements related to the events, on the other hand, the formed event graph also reflects the incidence relation among the events in the event graph, so that the follow-up query and analysis of the correlation among the events are facilitated, and the accuracy and the recall rate of the event query can be effectively improved.
The event elements described above may include entities, attributes, and event characteristics of the event. These event elements can be obtained by information extraction of the event description information. In the process of information extraction, the entity, the attribute and the event feature can be subjected to deduplication processing, namely, substantially the same event element can correspond to the same node in the event map, so that the data recorded in the event map is more accurate and effective.
The specific deduplication method may vary from event element to event element. For the entities involved in the event, a manner of entity alignment can be adopted, and all the entities are expressed by adopting a unified standard. And for the event features, a text clustering mode can be adopted, and the event features of which the distance of the text vector is smaller than a preset threshold value are assigned to the same category identification.
The event information processing method of the embodiment of the invention can be realized by a data service platform for providing storage and query services of event information. The event information may be event information for a certain field, for example, event information for processing news events, events generated for sales of e-commerce platforms, alarm events for website operation systems, and the like, and the event information source may be from public information collection for each website, event documents uploaded by users, event logs provided by some system monitoring platforms, and the like. The event information can be transmitted to a data service platform in a text form, and an event map is formed after processing. The data service platform may provide query services to the user based on the generated event graph. In the process of inquiring, a user inputs an inquiring condition and then converts the inquiring condition into a map inquiring language by a data service platform, thereby executing retrieval based on a map structure and returning event content meeting the inquiring condition to the user.
The technical solution of the present invention is further illustrated by some specific examples.
Example one
As shown in fig. 1, which is one of the flow diagrams of an event information processing method according to an embodiment of the present invention, the method may be applied to a data service platform that provides a storage and query service of event information, and the method includes:
s101: the information extraction is carried out on the event description information of the event, and a plurality of event elements of the event and a first incidence relation among the event elements are obtained. An event as referred to herein may be a basic element event comprising an execution subject, a time and/or a place. The number of events may be one or more. The event description information can exist in a text form, core elements forming the event are obtained by extracting the information of the event description information, and text contents which have no practical significance or have small information quantity are removed. These event elements may include: entities, attributes, and event characteristics of an event. The information extraction processing is mainly to extract information such as specific category entities, attributes, relationships and the like from a given text, so that the unstructured text is structured, and subsequent statistical analysis processing is facilitated.
Where an entity corresponds to an object that actually exists in the real world, such as a person, a building, an item, etc., and the attribute is some information attached to the entity, such as the identity of the person, the gender, the longitude and latitude coordinates of the building, etc. Event features are generally text segments relating to the nature or character of an event, and such text segments often represent abstract features of an event. For example, a news event description can embody a text segment of a news type (e.g., entertainment news, economic news), and for example, a consumption event of an e-commerce platform, a text segment of a commodity type (e.g., electronic products, food) purchased by a customer, and the like. The first association relationship may include an association relationship between the entity and the attribute and/or an association relationship between the entities.
The information extraction process described above may involve two processes:
one is the calibration of text segments, entities (e.g., names of people, places, etc.), attributes (e.g., identification, phone numbers, latitude and longitude, etc.), and event characteristics. Namely, whether each text segment belongs to an entity, an attribute or an event characteristic is marked, so that subsequent duplication removal and further processing for forming an event map are facilitated. For the calibration of the text segment, the extraction can be performed through a sequence labeling model, the sequence labeling model predicts a label for each character of an input sentence to indicate whether the character is an entity word, for example, if the result "O PER" obtained by inputting "human name" is input, the "PER" indicates that the character is a human name entity, so that the "name" can be known.
The other is the extraction of the relationship information, i.e., the aforementioned first association relationship. Such a relationship may be a relationship between an entity and an attribute, or a relationship between an entity and an entity. The correspondence of attributes to entities is, for example, "person-mobile number", i.e., the mobile number corresponds to the person. Entity to entity relationships such as "XX building belongs to XX city". This type of information is extracted using a relational extraction model: the relationship extraction model classifies a given sentence and a specified entity attribute pair, such as a given "Li Ming" (330220198912211234, 13505121232), a specified entity "Li Ming", an attribute "13505121232", and a model giving a relationship between entities and attributes "communication method".
Through the information extraction processing, valuable element information in the event description information is extracted, and calibration is carried out according to the entity, the attribute, the event characteristic and the relationship information, so that subsequent processing can be carried out.
In the embodiment of the invention, in the information extraction process, the deduplication processing can be further executed. The specific duplication removing mode can be different according to different event elements, and the duplication removing is mainly carried out in a mode of entity alignment and text clustering.
Entities involved in an event can be processed in an entity alignment mode, that is, an entity of an event is aligned with entities of other events, so that the same entities have the same entity name, and the same entities can correspond to the same node in an event graph. Due to different event description modes, entities can be expressed differently in different same event description texts, which brings difficulty to the analysis of events, for example, the occurrence place of event a is an "AABB route", the occurrence place of event B is a "BBAA route", and the two routes are actually the same route, and are only the difference between new and old road names. And if the same name appears in different events, whether the same person exists needs to be judged. For the address entities, the address entities can be verified by an address standardization technology through a third-party geographic information database, and different addresses are converted into the same standardized expression, so that alignment is realized. And for the name entity, the standard name is obtained by comparing the associated attributes, such as the judgment of an identity card, a mobile phone number, a bank card and the like. For other entities such as organization names, the text similarity and the context similarity of the names can be comprehensively used for judgment and determination. After the entities of a plurality of events are aligned to form a uniform expression, the entities can correspond to the same node in the event graph generated subsequently, so that an accurate incidence relation is established among all the incidents.
The event features can be clustered by adopting a text clustering mode: the event feature of the event and the event features of other events are subjected to text clustering, specifically, the event feature of the event is represented as a first vector, the event features of the other events are represented as a second vector, the distance between the first vector and the second vector is calculated, and if the distance of the text vector is smaller than a preset threshold value, the same category identification is given to the event feature of the event and the event features of the other events, so that the event features which are substantially the same or very close to each other correspond to the same node in the event graph. In particular, for text segments such as event features, since it is difficult to align like entities, deduplication processing is performed by means of a text clustering technique. The text segments of the event features are expressed as text vectors, specifically, the text vectors can be high-dimensional sparse vectors based on a vector space model, and also can be low-dimensional dense vectors based on deep representation learning. And calculating the distance between the text vectors, clustering the texts by using a clustering algorithm (such as a K-means algorithm), and assigning the same category identification to the same category for clustering. For example, for the following four text segments related to news content, "XX star participates in awards show", "XX tv drama starts", "XX falls at 1000 points", "XX company falls into financial crisis", two categories can be clustered, the first two categories are entertainment news categories, and the last two categories are finance news categories. It should be noted that the category in the text clustering process may be determined according to actual requirements, such as the category determined by the text segment based on the clustering requirements of the news event in the previous example.
S102: and giving an event identifier to the event, and establishing a second association relationship between all or part of event elements of the event and the event identifier. In order to embody the concept of events in the event graph, each event may be assigned a unique event identifier, and the event identifier may correspond to a node in the event graph. In terms of logical relationship, the entity and the event feature related to the event have a direct association relationship with the event identifier, and the attribute may establish a direct association relationship with the event identifier or an indirect association relationship with the node corresponding to the entity.
S103: and constructing nodes in the event graph by using the event identifications and the event elements of the events, and constructing edges in the event graph by using the first incidence relation and the second incidence relation, thereby generating the event graph for the events. In the generated event graph, the same event element may correspond to the same node in the event graph. In this step, the processing of map construction is mainly involved. Specifically, the entity, the attribute, the event feature and the event identifier obtained in the previous step are all corresponding to one node in the event graph, then edge connection is established according to the incidence relation between the entity and between the entity and the attribute, and in addition, all the nodes corresponding to the entity and the event feature are connected with the node corresponding to the event identifier, so that the topological structure of the event graph is formed. The entity name, the attribute content, the category identifier of the event feature, and the event identifier are written into the storage areas corresponding to the respective nodes as the node content, and the association relationship between the entities and the attribute and/or the association relationship between the entities can be written into the storage areas corresponding to the edges as the content of the edges, for example, "communication mode" is written into the edges between the nodes corresponding to the mobile phone number (attribute) and the person name (entity). In the above process of forming the event map, the event map may be created from the beginning or the event information may be added on the basis of an existing event map, and the processing method is the same.
In the event graph, information of a plurality of events is included, and new event information may be added to the event graph. Based on the foregoing deduplication processing, the same entity, attribute, and event feature correspond to the same node in the event graph, so that different incidents generate interleaved association relationships, which can be more beneficial to the serial-parallel analysis of subsequent events.
Fig. 2 is a schematic diagram of an event graph structure according to an embodiment of the present invention. Two news events are shown by way of example, with the event identifications "event 1" and "event 2" being assigned. The event description text of event 1 is "a 2 company (trade name XXXA2) purchased a stock of a1 company (trade name XXXA1) listed in the XXX stock exchange market", and the event description text of event 2 is "a 1 company director plum XX (identification XXXX) purchased a XXX basketball team on a personal name".
After the description texts of the two news events are extracted, entity information including A1 company, A2 company, XXX stock trading market, Li XX and XXX basketball team is obtained, attribute information including industrial and commercial codes XXXA1, industrial and commercial codes XXXA2 and identity XXXXX are obtained, and the obtained association relationship between the entities and the attributes is as follows: the relationship between "Industrial and commercial code XXXA 1" and "A1 company" is "ID", and the relationship between "ID XXXX" and "Li XX" is "ID".
In addition, the text segment for which the event feature is obtained includes "stock purchased from A1 company marketed in the XXX stock exchange market" (referred to as event feature 1), "basketball team XXX purchased on a personal name" (referred to as event feature 2). The text clustering process is performed on the text segments of the event features, and it should be noted that the clustering process herein involves calculating distances between a plurality of text vectors in a vector space, that is, the text vectors corresponding to the event features 1 and 2 and other text vectors in the vector space perform a plurality of text vector distances, so as to determine the category to which the text vectors belong, where the vector space is a vector space formed by text vectors of event features in a large number of news events. After the clustering process, the obtained result is that the event characteristic 1 is clustered into news of "finance class", and the event characteristic 2 is clustered into news of "sports class" and "finance class", it should be noted that the same event can also be clustered into a plurality of different classes.
Based on the extraction result, a graph is constructed to form an event graph as shown in fig. 2, and a dotted circle in the graph corresponds to a node corresponding to the event identifier. As can be seen from the figure, although the event 1 and the event 2 are two different events, they are related to each other because they are designed to be the same company "a 1 company".
Based on the event map constructed above, query processing of events can be performed. As shown in fig. 3, which is a second flowchart of the event information processing method according to the embodiment of the present invention, the method includes:
s201: acquiring a query condition of a user, wherein the query condition comprises event elements and/or a first incidence relation between the event elements. Specifically, in this step, the obtaining of the query condition of the user may specifically include: receiving a query request of a user, wherein the query request comprises event description information of an event to be queried; and extracting information of the event description information, and acquiring event elements of the event to be inquired and/or a first incidence relation between the event elements as inquiry conditions. The event elements may include entities, attributes, and event features of the event, among others. After the query conditions are obtained, the query conditions can be converted into query language of the graph database for query processing.
S202: and traversing the event graph according to the query condition, and determining a first node and/or edge corresponding to the event element and/or the first incidence relation. For example, based on the event graph shown in fig. 2, the query condition is "related to a1 company" and "financial class event", and based on such query condition, the node corresponding to "a 1 company" and the node corresponding to "financial class", i.e., the first node, are determined from the event graph.
S203: and according to the first node and/or the first edge, determining a second node which is associated with the first node and/or the first edge and corresponds to the event identifier, and acquiring the event identifier from the second node. Based on the node determined in the previous step, according to the node corresponding to the event identifier connected to the two nodes, the nodes corresponding to the "event 1" and the "event 2", that is, the second node, can be determined, and then the corresponding event identifiers can be obtained.
S204: and acquiring corresponding event description information according to the event identifier, and returning the event description information to the user. After the event identifier is obtained, the description text of the news event corresponding to the event identifier can be obtained from the text database, and then returned to the user requesting for query.
According to the event information processing method, event elements and the incidence relation among the elements are extracted from the event description information, an event map is constructed based on the event elements, and the event information is stored by adopting a structured topological structure, so that the follow-up serial and parallel query and analysis of a large number of events are facilitated. In the process of inquiring the event, the user traverses the event graph based on the input inquiry condition, and can determine the related associated event based on the association relation between the nodes, thereby effectively improving the recall rate and the accuracy rate of the event inquiry.
Example two
As shown in fig. 4, which is a schematic structural diagram of an event information processing apparatus according to an embodiment of the present invention, the apparatus may be disposed in a data service platform that provides a storage and query service of event information, and the apparatus includes:
the information extraction module 11 is configured to extract information from event description information of an event, and acquire a plurality of event elements of the event and a first association relationship between the event elements. The event elements may include entities, attributes, and event features of the event, among others. The information extraction processing is mainly to extract information such as specific category entities, attributes, relationships and the like from a given text, so that the unstructured text is structured, and subsequent statistical analysis processing is facilitated. The information extraction process described above may involve two processes: one is the calibration of text segments, entities (e.g., names of people, places, etc.), attributes (e.g., identification, phone numbers, latitude and longitude, etc.), and event characteristics. Namely, whether each text segment belongs to an entity, an attribute or an event characteristic is marked, so that subsequent duplication removal and further processing for forming an event map are facilitated. For the calibration of the text segment, the extraction can be performed through a sequence labeling model. The other is the extraction of the relationship information, i.e., the aforementioned first association relationship. The relationship can be the relationship between the entity and the attribute, or the relationship between the entity and the entity, and the information of the type is extracted by using a relationship extraction model.
And the event identifier processing module 12 is configured to assign an event identifier to the event, and establish a second association relationship between all or part of event elements of the event and the event identifier. In order to embody the concept of events in the event graph, each event may be assigned a unique event identifier, and the event identifier may correspond to a node in the event graph. In terms of logical relationship, the entity and the event feature related to the event have a direct association relationship with the event identifier, and the attribute may establish a direct association relationship with the event identifier or an indirect association relationship with the node corresponding to the entity.
And the graph generating module 13 is configured to construct nodes in the event graph by using the event identifiers and the event elements of the events, and construct edges in the event graph by using the first incidence relation and the second incidence relation, so as to generate the event graph for the events. Wherein, in the generated event graph, the same event element may correspond to the same node in the event graph. Specifically, the entity, the attribute, the event feature and the event identifier which are obtained in the past are all corresponding to one node in the event graph, then edge connection is established according to the incidence relation between the entity and between the entity and the attribute, in addition, all the nodes corresponding to the entity and the event feature are connected with the node corresponding to the event identifier, and therefore the topological structure of the event graph is formed. The entity name, the attribute content, the category identifier of the event feature, and the event identifier are written into the storage area corresponding to each node as the node content, and the association relationship between the entity and the attribute and/or the association relationship between the entities can be written into the storage area corresponding to the edge as the content of the edge. The above-mentioned process of forming the event map may be to establish the event map from the beginning or to add event information on the basis of an existing event map, and the processing mode is the same.
In the embodiment of the invention, in the information extraction process, the deduplication processing can be further executed. The specific duplication removing mode can be different according to different event elements, and the duplication removing is mainly carried out in a mode of entity alignment and text clustering.
In order to perform the process of entity alignment, the apparatus may further include: and the entity alignment module 14 is configured to perform entity alignment processing on the entity of the event and the entities of other events, so that the same entity has the same entity name, and thus the same entity may correspond to the same node in the event graph.
In order to perform the text clustering process, the apparatus may further include: the text clustering module 15 is configured to perform text clustering on the event features of the event and the event features of other events, specifically, represent the event features of the event as a first vector, represent the event features of other events as a second vector, calculate a distance between the first vector and the second vector, and if the distance of the text vector is smaller than a preset threshold, assign the same category identifier to the event features of the event and the event features of other events, so that the substantially same or very close event features correspond to the same node in the event graph. In the event graph, information of a plurality of events is included, and new event information may be added to the event graph. Based on the foregoing deduplication processing, the same entity, attribute, and event feature correspond to the same node in the event graph, so that different incidents generate interleaved association relationships, which can be more beneficial to the serial-parallel analysis of subsequent events.
Based on the event map constructed above, query processing of events can be performed. As shown in fig. 5, which is a second schematic structural diagram of an event information processing apparatus according to an embodiment of the present invention, the apparatus may be disposed in a data service platform that provides storage and query services for event information, and the apparatus includes:
the query condition obtaining module 21 is configured to obtain a query condition of a user, where the query condition includes event elements and/or a first association relationship between event elements. Specifically, the obtaining of the query condition of the user may specifically include: receiving a query request of a user, wherein the query request comprises event description information of an event to be queried; and extracting information of the event description information, and acquiring event elements of the event to be inquired and/or a first incidence relation between the event elements as inquiry conditions. The event elements may include entities, attributes, and event features of the event, among others. After the query conditions are obtained, the query conditions can be converted into query language of the graph database for query processing.
And the traversal query module 22 is configured to traverse the event graph according to the query condition, and determine a first node and/or an edge corresponding to the event element and/or the first association relationship.
And the event identifier determining module 23 is configured to determine, according to the first node and/or the first edge, a second node associated with the first node and/or the first edge and corresponding to the event identifier, and acquire the event identifier from the second node.
And the query result feedback module 24 is configured to obtain corresponding event description information according to the event identifier, and return the event description information to the user. After the event identifier is obtained, the description text of the news event corresponding to the event identifier can be obtained from the text database, and then returned to the user requesting for query.
The detailed description of the above processing procedure, the detailed description of the technical principle, and the detailed analysis of the technical effect are described in the foregoing embodiments, and are not repeated herein.
According to the event information processing device, the event elements and the incidence relation among the elements are extracted from the event description information, the event graph is constructed based on the event elements, and the event information is stored by adopting the structured topological structure, so that the follow-up serial and parallel query and analysis of a large number of events are facilitated. In the process of inquiring the event, the user traverses the event graph based on the input inquiry condition, and can determine the related associated event based on the association relation between the nodes, thereby effectively improving the recall rate and the accuracy rate of the event inquiry.
EXAMPLE III
The foregoing embodiment describes a processing flow of event information and a device structure, and the functions of the method and the device can be implemented by an electronic device, as shown in fig. 6, which is a schematic structural diagram of the electronic device according to an embodiment of the present invention, and specifically includes: a memory 110 and a processor 120.
And a memory 110 for storing a program.
In addition to the programs described above, the memory 110 may also be configured to store other various data to support operations on the electronic device. Examples of such data include instructions for any application or method operating on the electronic device, contact data, phonebook data, messages, pictures, videos, and so forth.
The memory 110 may be implemented by any type or combination of volatile or non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disks.
The processor 120, coupled to the memory 110, is used for executing the program in the memory 110 to perform the operation steps of the processing method of the event information described in the foregoing embodiments.
Further, the processor 120 may also include various modules described in the foregoing embodiments to perform the processing of the event information, and the memory 110 may be used, for example, to store data required by the modules to perform operations and/or output data.
The detailed description of the above processing procedure, the detailed description of the technical principle, and the detailed analysis of the technical effect are described in the foregoing embodiments, and are not repeated herein.
Further, as shown, the electronic device may further include: communication components 130, power components 140, audio components 150, display 160, and other components. Only some of the components are schematically shown in the figure and it is not meant that the electronic device comprises only the components shown in the figure.
The communication component 130 is configured to facilitate wired or wireless communication between the electronic device and other devices. The electronic device may access a wireless network based on a communication standard, such as WiFi, a mobile communication network, such as 2G, 3G, 4G/LTE, 5G, or a combination thereof. In an exemplary embodiment, the communication component 130 receives a broadcast signal or broadcast related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 130 further includes a Near Field Communication (NFC) module to facilitate short-range communications. For example, the NFC module may be implemented based on Radio Frequency Identification (RFID) technology, infrared data association (IrDA) technology, Ultra Wideband (UWB) technology, Bluetooth (BT) technology, and other technologies.
The power supply component 140 provides power to the various components of the electronic device. The power components 140 may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power for an electronic device.
The audio component 150 is configured to output and/or input audio signals. For example, the audio assembly 150 includes a Microphone (MIC) configured to receive external audio signals when the electronic device is in an operational mode, such as a call mode, a recording mode, and a voice recognition mode. The received audio signal may further be stored in the memory 110 or transmitted via the communication component 130. In some embodiments, audio assembly 150 also includes a speaker for outputting audio signals.
The display 160 includes a screen, which may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive an input signal from a user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundary of a touch or slide action, but also detect the duration and pressure associated with the touch or slide operation.
Those of ordinary skill in the art will understand that: all or a portion of the steps of implementing the above-described method embodiments may be performed by hardware associated with program instructions. The aforementioned program may be stored in a computer-readable storage medium. When executed, the program performs steps comprising the method embodiments described above; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.
Claims (16)
1. A method for processing event information comprises the following steps:
extracting information of event description information of an event, and acquiring a plurality of event elements of the event and a first incidence relation among the event elements;
giving an event identifier to the event, and establishing a second association relationship between all or part of event elements of the event and the event identifier; and constructing nodes in an event graph by using the event identification and the event elements of the event, constructing edges in the event graph by using the first incidence relation and the second incidence relation, and generating the event graph for the event.
2. The method of claim 1, wherein the event elements include entities, attributes, and event features of an event.
3. The method of claim 2, further comprising:
and carrying out entity alignment processing on the entity of the event and the entities of other events, wherein the same entities have the same entity names.
4. The method of claim 2, further comprising:
performing text clustering on the event characteristics of the event and the event characteristics of other events:
representing event features of the event as a first vector;
representing event features of other events as a second vector;
calculating a distance between the first vector and a second vector;
and if the distance is smaller than a preset threshold value, giving the same category identification to the event characteristics of the event and the event characteristics of other events.
5. The method of claim 2, wherein the first association comprises an association between an entity and an attribute and/or an association between entities; the second incidence relation comprises incidence relation between entity and event characteristics and the event identification.
6. The method of claim 5, further comprising:
and recording the first incidence relation and the second incidence relation in a storage area corresponding to the edge of the event map.
7. A method for processing event information comprises the following steps:
acquiring a query condition of a user, wherein the query condition comprises event elements and/or a first incidence relation among the event elements;
traversing an event graph according to the query condition, and determining a first node and/or edge corresponding to the event element and/or the first incidence relation;
determining a second node which is associated with the first node and/or the first edge and corresponds to the event identifier according to the first node and/or the first edge, and acquiring the event identifier from the second node;
and acquiring corresponding event description information according to the event identifier, and returning the event description information to the user.
8. The method of claim 7, wherein the event elements include entities, attributes, and event features of an event.
9. The method of claim 7, wherein obtaining query conditions for a user comprises:
receiving a query request of a user, wherein the query request comprises event description information of an event to be queried;
and extracting information of the event description information, and acquiring event elements of the event to be queried and/or a first incidence relation between the event elements as the query condition.
10. An apparatus for processing event information, comprising:
the information extraction module is used for extracting information of event description information of an event and acquiring a plurality of event elements of the event and a first incidence relation among the event elements;
the event identifier processing module is used for endowing the event with an event identifier and establishing a second association relationship between all or part of event elements of the event and the event identifier;
and the graph generation module is used for constructing nodes in the event graph by using the event identifications and the event elements of the events, constructing edges in the event graph by using the first incidence relation and the second incidence relation, and generating the event graph for the events.
11. The apparatus of claim 10, wherein the event elements include entities, attributes, and event features of an event.
12. The apparatus of claim 11, further comprising:
and the entity alignment module is used for carrying out entity alignment processing on the entity of the event and the entities of other events, and the same entities have the same entity names.
13. The apparatus of claim 11, further comprising:
and the text clustering module is used for carrying out entity alignment processing on the entity of the event and the entities of other events, wherein the same entities have the same entity names.
14. An apparatus for processing event information, comprising:
the query condition acquisition module is used for acquiring a query condition of a user, wherein the query condition comprises event elements and/or a first incidence relation among the event elements;
the traversal query module is used for traversing the event graph according to the query condition and determining a first node and/or edge corresponding to the event element and/or the first incidence relation;
an event identifier determining module, configured to determine, according to the first node and/or the first edge, a second node associated with the first node and/or the first edge and corresponding to an event identifier, and obtain the event identifier from the second node;
and the query result feedback module is used for acquiring corresponding event description information according to the event identifier and returning the event description information to the user.
15. An electronic device, comprising:
a memory for storing a program;
a processor for executing the program stored in the memory to perform the event information processing method of any one of claims 1 to 6.
16. An electronic device, comprising:
a memory for storing a program;
a processor for executing the program stored in the memory to perform the event information processing method according to any one of claims 7 to 9.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010603795.0A CN113934764A (en) | 2020-06-29 | 2020-06-29 | Event information processing method and device and electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010603795.0A CN113934764A (en) | 2020-06-29 | 2020-06-29 | Event information processing method and device and electronic equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113934764A true CN113934764A (en) | 2022-01-14 |
Family
ID=79272832
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010603795.0A Pending CN113934764A (en) | 2020-06-29 | 2020-06-29 | Event information processing method and device and electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113934764A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114595997A (en) * | 2022-03-21 | 2022-06-07 | 联想(北京)有限公司 | Data processing method and device and electronic equipment |
-
2020
- 2020-06-29 CN CN202010603795.0A patent/CN113934764A/en active Pending
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114595997A (en) * | 2022-03-21 | 2022-06-07 | 联想(北京)有限公司 | Data processing method and device and electronic equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108021929B (en) | Big data-based mobile terminal e-commerce user portrait establishing and analyzing method and system | |
US8224766B2 (en) | Comparing spatial-temporal trails in location analytics | |
KR101419504B1 (en) | System and method providing a suited shopping information by analyzing the propensity of an user | |
CN105095211B (en) | The acquisition methods and device of multi-medium data | |
TWI703862B (en) | Content recommendation method and device | |
CN110019616B (en) | POI (Point of interest) situation acquisition method and equipment, storage medium and server thereof | |
CA3036664A1 (en) | Method for data structure relationship detection | |
CN113127506B (en) | Target query statement construction method and device, storage medium and electronic device | |
CN110543477B (en) | Label construction system and method | |
CN116739836B (en) | Restaurant data analysis method and system based on knowledge graph | |
CN112597775A (en) | Credit risk prediction method and device | |
WO2016188334A1 (en) | Method and device for processing application access data | |
CN108512674B (en) | Method, device and equipment for outputting information | |
CN113934764A (en) | Event information processing method and device and electronic equipment | |
CN112363996A (en) | Method, system, and medium for building a physical model of a power grid knowledge graph | |
CN116703515A (en) | Recommendation method and device based on artificial intelligence, computer equipment and storage medium | |
US20220327147A1 (en) | Method for updating information of point of interest, electronic device and storage medium | |
CN113392920B (en) | Method, apparatus, device, medium, and program product for generating cheating prediction model | |
CN110062112A (en) | Data processing method, device, equipment and computer readable storage medium | |
US20150073902A1 (en) | Financial Transaction Analytics | |
CN112085566B (en) | Product recommendation method and device based on intelligent decision and computer equipment | |
CN113780318B (en) | Method, device, server and medium for generating prompt information | |
CN114153860A (en) | Business data management method and device, electronic equipment and storage medium | |
CN114897571A (en) | Method, system and storage medium for processing goods data of shopping mall based on private domain e-commerce platform | |
KR20220074571A (en) | A method and an apparatus for embedding sns contents based on deep learning analysis for collecting channel information of marketing knowledge graphs |
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 |