CN111475612A - Construction method, device and equipment of early warning event map and storage medium - Google Patents
Construction method, device and equipment of early warning event map and storage medium Download PDFInfo
- Publication number
- CN111475612A CN111475612A CN202010136834.0A CN202010136834A CN111475612A CN 111475612 A CN111475612 A CN 111475612A CN 202010136834 A CN202010136834 A CN 202010136834A CN 111475612 A CN111475612 A CN 111475612A
- Authority
- CN
- China
- Prior art keywords
- event
- information
- event information
- map
- text
- 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
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3344—Query execution using natural language analysis
-
- 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
- G06F16/34—Browsing; Visualisation therefor
- G06F16/345—Summarisation for human users
-
- 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
- G06F16/35—Clustering; Classification
-
- 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
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9536—Search customisation based on social or collaborative filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
Abstract
The application discloses a construction method of an early warning event map, which relates to the field of big data, and comprises the following steps: when an event map construction instruction is received, acquiring text information for constructing the event map; event acquisition is carried out on the text information, and at least one piece of first event information corresponding to the text information and an early warning label corresponding to the at least one piece of first event information are obtained; screening the at least one first event information to obtain at least one second event information for constructing an event map; determining a hierarchical relationship between the at least one second event information according to the feature information contained in the at least one second event; and storing the at least one piece of second event information in a corresponding graph database according to the hierarchical relationship so as to complete the construction of the event graph. The application also provides a device, computer equipment and a storage medium. The quality of the event map is improved when the event map is constructed.
Description
Technical Field
The application relates to the technical field of big data, in particular to a construction method and device of an early warning event map, computer equipment and a storage medium.
Background
With the rapid popularization of the internet, various social network media, such as blogs, wiki, podcasts, forums, social networks, content communities and the like, are developed vigorously, and become the most important channels for people to publish, obtain and transmit event information, and the arrangement and research of the events are helpful for people to understand the event development rules and guide production and life, and have wide needs and applications in the fields of news recommendation, public opinion analysis and the like.
At present, an event graph is a new concept, and records an association relationship and a causal relationship between events, in the event graph, an effect event corresponding to each event is different, and different events have different influences on the development of the whole event. The event map is also a knowledge map, and the knowledge map is a modern theory which achieves the aim of multi-discipline fusion by combining theories and methods of applying disciplines such as mathematics, graphics, information visualization technology, information science and the like with methods such as metrology introduction analysis, co-occurrence analysis and the like and utilizing the visualized map to vividly display the core structure, development history, frontier field and overall knowledge architecture of the disciplines.
The event map is different from the entity knowledge map taking a concept-actual example as a core, so that the event extraction effect is difficult to control, an event extraction system is difficult to distinguish which events are important and which events are unimportant, and meanwhile, the events are not accurately checked, so that the quality of the constructed event map is low.
Disclosure of Invention
The application provides a construction method and device of an early warning event map, computer equipment and a storage medium, so as to improve the purity of the event map and the construction quality of the event map.
In a first aspect, the present application provides a method for constructing an early warning event map, where the method includes:
when an event map construction instruction is received, acquiring text information for constructing the event map;
event acquisition is carried out on the text information, and at least one piece of first event information corresponding to the text information and an early warning label corresponding to the at least one piece of first event information are obtained;
screening the at least one first event information to obtain at least one second event information for constructing an event map;
determining a hierarchical relationship between the at least one second event information according to the feature information contained in the at least one second event;
and storing the at least one piece of second event information in a corresponding graph database according to the hierarchical relationship so as to complete the construction of the event graph.
In a second aspect, the present application further provides an early warning event map constructing apparatus, where the apparatus includes:
the text acquisition module is used for acquiring text information for constructing the event map when receiving an event map construction instruction;
the first processing module is used for acquiring an event from the text information based on an unsupervised text summarization technology to obtain at least one piece of first event information corresponding to the text information and an early warning label corresponding to the at least one piece of first event information;
the second processing module is used for screening the at least one first event information to obtain at least one second event information for constructing an event map;
the relationship determination module is used for determining the hierarchical relationship among the at least one piece of second event information according to the characteristic information contained in the at least one piece of second event information;
and the map construction module is used for storing the at least one piece of second event information in a corresponding map database according to the hierarchical relationship so as to complete the construction of the event map.
In a third aspect, the present application further provides a computer device comprising a memory and a processor; the memory is used for storing a computer program; the processor is used for executing the computer program and realizing the construction method of the early warning event map when the computer program is executed.
In a fourth aspect, the present application further provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the computer program causes the processor to implement the method for constructing the warning event map.
The application discloses a construction method, a construction device, computer equipment and a storage medium of an early warning event map, wherein when the event map needing risk early warning is constructed, an input event map construction instruction is received, text information for constructing the event map is acquired at the same time, then the acquired text information is acquired by using an unsupervised text summarization technology to obtain corresponding first event information, then the first event information is processed, including screening and duplicate removal, to obtain second event information for constructing the event map, and finally the second event information is stored in a corresponding map database to complete construction of the event map. The method and the device realize that the purity of the event map is improved when the event map is constructed, reduce the noise data in the obtained event map and improve the quality of the event map.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
Fig. 1 is a schematic flow chart of a method for constructing an early warning event map according to an embodiment of the present application;
FIG. 2 is a flowchart illustrating a step of obtaining second event information according to an embodiment of the present application;
FIG. 3 is a flowchart illustrating a step of processing first event information after time stamping according to an embodiment of the present application;
FIG. 4 is a schematic block diagram of an early warning event graph constructing apparatus according to an embodiment of the present application;
FIG. 5 is a block diagram showing a schematic configuration of a computer device according to an embodiment.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some, but not all, embodiments of the present application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
The flow diagrams depicted in the figures are merely illustrative and do not necessarily include all of the elements and operations/steps, nor do they necessarily have to be performed in the order depicted. For example, some operations/steps may be decomposed, combined or partially combined, so that the actual execution sequence may be changed according to the actual situation.
It is to be understood that the terminology used in the description of the present application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in the specification of the present application and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It should also be understood that the term "and/or" as used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.
Some embodiments of the present application will be described in detail below with reference to the accompanying drawings. The embodiments described below and the features of the embodiments can be combined with each other without conflict.
Referring to fig. 1, fig. 1 is a schematic flow chart illustrating a method for constructing an early warning event map according to an embodiment of the present application.
Specifically, the method comprises the following steps:
and step S101, when an event map construction instruction is received, acquiring text information for constructing the event map.
When the event map of risk early warning is constructed, or text information required by the construction of the event map is carried out, the obtained text information is related records of things which happen in the past, and the event information recorded in the text information is correspondingly processed, so that a more reasonable means can be obtained when the same or similar events occur, and a better effect is obtained.
Specifically, when an event map construction instruction is received, input text information for performing event map construction is received, and the event map is constructed according to the received text information. The text information is related to a scene constructed according to an event map, when risk early warning is carried out, the constructed event map and a current event are utilized to carry out corresponding result forenotice when risk forenotice is required, related personnel can be timely reminded when the result is a risk result, and then risk can be avoided through corresponding processing operation.
The risk early warning method has various scenes, for example, for the risk early warning of financial events, the triggering factors of the financial events are determined and the generation of the triggering factors is avoided through the analysis and the processing of the past financial events, and when a certain operation is performed, the positive effect is generated more effectively.
For another example, for the prediction of an event result, for an event, the occurrence of a small event will affect the occurrence of the whole event, so that the occurrence of bad things can be avoided through risk pre-warning.
In some embodiments, the obtained text information for constructing the event graph is not limited, and may specifically include: when an event map construction instruction is received, receiving uploaded text information for constructing an event map; or when an event map building instruction is received, reading the input text link, and acquiring text information for building the event map according to the text link.
Specifically, the source of the text information for constructing the event graph may be diversified, for example, the user may upload the corresponding text information, and may also obtain the corresponding text information on a network or other platform, specifically, the obtained text information is related to the subject information of the currently constructed event graph according to the attribute of the event graph that needs to be constructed.
Step S102, event acquisition is carried out on the text information, and an early warning label corresponding to at least one first event information level corresponding to the text information is obtained.
After the text information for constructing the event graph is acquired, event information is acquired, and specifically, event information included in the text information may be acquired by using an unsupervised text summarization technology to obtain a plurality of first event information corresponding to the text information, and different early warning labels corresponding to different event information are determined according to results corresponding to different event information, where the first event information is only used for distinguishing other event information and represents event information at different processing stages, and the early warning labels are classified into different grades and described by using different number numbers or characters.
When event information included in text information is acquired, the event information is acquired as long as there is an event, and the event information is acquired regardless of the size of the event, whether the event is repeatedly acquired, or the like.
In some embodiments, when the text information is processed to obtain the corresponding first event information, the method includes: performing word segmentation processing on the text information to obtain a plurality of word units; and constructing a graph model, and processing the word units based on the graph model to obtain at least one piece of first event information corresponding to the text information and an early warning label corresponding to the at least one piece of first event information.
After receiving the text information for event map construction, extracting the text information for event information by using an unsupervised text summarization technology to obtain corresponding first event information.
Specifically, after receiving input text information, processing the text information by using a TextRank algorithm, dividing the text information into a plurality of word units, wherein the word units comprise words and sentences, then constructing a corresponding graph model, and sequencing the obtained word units based on the graph model to obtain word units with higher importance in the text information, so as to obtain first event information corresponding to the text information.
When text information is processed by using a TextRank algorithm, the textRank algorithm is used for acquiring keyword information and/or keyword group information contained in the text information, a constructed graph model is used for recording information, counting the occurrence frequency of word units, filtering the word units with extremely low occurrence frequency or lower than a certain threshold value, storing the corresponding word units only when the occurrence frequency of the word units meets a certain requirement, and further obtaining a plurality of first event information corresponding to the text information according to the relation between the word units.
In addition, when the graph model is constructed to process the word units, the weight values of all the word units can be calculated, effective word units in the text information are further determined through the weight values, and then a plurality of corresponding first event information are obtained according to the effective word units and the relations among the word units.
Step S103, screening the at least one first event information to obtain at least one second event information for constructing an event map.
After the obtained text information is processed by using an unsupervised text summarization technology to obtain first event information, the first event information is processed for the second time to obtain second event information for constructing an event map.
Since all event information contained in the text information is acquired when the first event information is obtained, there are cases where events are incomplete or repeated, and the like in the event information contained in the obtained first event information, and therefore, after the first event information is obtained, the first event information needs to be subjected to secondary processing to obtain second event information, so that the obtained second event information can be used for accurately constructing an event graph.
Step S104, determining the hierarchical relationship between the at least one second time information according to the feature information contained in the at least one second event.
After the event information corresponding to the text information is obtained, the association relationship between the event information can be determined, which mainly includes the hierarchical relationship. After screening the at least one first event information to obtain at least one second event information, firstly obtaining characteristic information contained in the at least one second event information, wherein the characteristic information is used for determining the hierarchical relationship among the event information, so as to determine the hierarchical relationship among the at least one second event according to the obtained characteristic information.
When the feature information included in the event information is obtained, by identifying the keyword information in the event information and the incidence relation among the keywords, for example, sports, basketball, football, table tennis and the like are all keywords, and meanwhile, sports belongs to the upper level of the basketball, the football and the table tennis, that is, a certain incidence relation exists.
A certain hierarchical relationship exists among any things, the hierarchical relationship among events is determined through keywords contained in each event information, taking sports as an example, the sports generally comprises football, basketball, tennis, racing car, diving, ping-pong and the like, and a certain hierarchical relationship also exists for events, for example, the sports events comprise match events, competition events and the like, and the hierarchical relationship or the association relationship before the events is determined, so that the hierarchy of the obtained time map is more obvious, and the association among the events is clearer.
And S105, storing the at least one piece of second event information in a corresponding graph database according to the hierarchical relationship so as to complete the construction of the event graph.
And after the first event information is processed to obtain second event information for constructing the event map, storing the second event information in a corresponding map database to complete construction of the event map for risk early warning.
Among them, commonly used graph databases include Neo4j, FlockDB, allegrograph, graph db, infinitiegraph, and the like, and the most advanced graph database Neo4j is used in this embodiment.
Specifically, because a certain hierarchical relationship exists between the event information, when storing, the label of each node is determined according to the relationship between the event information, and at the same time, attribute information associated with the event information, such as time information of the event information, event category information, and the like, is recorded on each node.
In some embodiments, when the building of the event graph is completed by using the second event information, the method includes: acquiring a hierarchy label corresponding to the second event information, and determining a storage hierarchy of the second event according to the hierarchy label; and storing the second event information according to the storage hierarchy, and establishing an incidence relation between the second event information to complete the construction of the event map.
The hierarchy corresponds to the nodes, if the hierarchical relationship among the events has 3 layers, the 3 layers respectively correspond to one node, the nodes are marked by using the feature description of the event information and the hierarchy labels to obtain corresponding labels, and the hierarchy information of different hierarchy labels is different. For example, the first layer event information is sports, the second layer event information is sports, the third layer event information is football, and a certain hierarchical relationship exists among the first layer event information, the second layer event information, the third layer event information and the football: the first level is sports, the second level is sports and the third level is football. Then the corresponding 3 nodes are the first node sports, the second node sports and the third node football, respectively. In the event graph, there are several layers of hierarchical relationships between the event information, and then there are several layers of node relationships.
In the above-described method for constructing an event graph for risk early warning, when an event graph for risk early warning is to be constructed, an input event graph construction instruction is received, text information for constructing the event graph is acquired at the same time, then the acquired text information is acquired by using an unsupervised text summarization technology to obtain corresponding first event information, then the first event information is processed, including screening and deduplication, to obtain second event information for constructing the event graph, and finally the second event information is stored in a corresponding graph database to complete construction of the event graph. The method and the device realize that the purity of the event map is improved when the event map is constructed, reduce the noise data in the obtained event map and improve the quality of the event map.
Further, referring to fig. 2, fig. 2 is a flowchart illustrating a step of obtaining second event information in an embodiment of the present application.
Specifically, step S103, processing the first text information to obtain second event information for constructing an event graph, including:
step S201, obtaining time information corresponding to the first event information, so as to perform time marking on the first event information by using the time information.
Step S202, screening the first event information subjected to the time marking to obtain second event information subjected to the event map construction.
When the first event information is processed, the first event information is firstly subjected to information perfection, and then the event information after the information perfection is processed. In the information processing method, when information is processed at night, the integrity of an event is not improved, for example, information recorded by certain obtained event information is incomplete, the integrity of the event information is not improved at this time, and specifically, after the first event information is obtained, the time information of the obtained first event information is improved according to the related information of the event information or the related information of the text information, and the event occurrence time and the like corresponding to the first event information are accurately determined.
In some embodiments, after the corresponding event information is obtained according to the text information, the time information corresponding to the event information needs to be accurately determined, and then the time information and the corresponding time information are associated to obtain the event information containing the time information.
For an event, there are many ways to represent time information of the event, so that a specific time, such as 8/2008, can be accurately represented by a description of the time, and words containing time information, such as yesterday, today, and days ago, can be indirectly represented by the description of the time.
For the indirect time description, the actual time information of the event needs to be determined according to the recorded event of the text information. Since there are many words or recording manners actually representing time information, it is necessary to determine the time information of each event information.
In some embodiments, when determining the time information corresponding to the event information, the following steps are included: when the event information contains the time information, acquiring the time information contained in the event information, and associating the acquired time information with the event information; when the event information does not contain the time information, acquiring the recording time information of the text information corresponding to the event information and the characteristic words of the event information, wherein the word attributes belong to the time attributes, and determining the time information corresponding to the event information according to the obtained recording time information and the characteristic words. And further correlating the obtained time information with the event information to obtain the event information containing the time information.
Since the description of the time is divided into a direct time description and an indirect time description (i.e., an indirect time description), when determining the time information of the event information, the time information included in the event information is first acquired, where the time information is divided into direct time information and indirect time information, when determining that the time information included in the event information is a direct time description, such as 8.8.2008, the obtained direct time description is directly used to mark the event information, and when determining that the time information included in the event information is an indirect time description, the corresponding processing is performed to obtain time information that can be used to accurately determine the event information. For example, for today's event information, if the event information is "stock is going up due to 5G development demand, yesterday a company" and today is 10/2020, then the event information corresponding to this event is 10/9/10/2020.
After the first event information marked by using time is obtained, secondary processing is carried out, and then second event information for constructing an event map is obtained. When the first event information after the time marking is processed, the event information is screened and deduplicated, the event information with incomplete event records is filtered, and repeated events are deduplicated to obtain second event information with complete event records and no repetition. By screening and de-duplicating the event information, the useless events and repeated events in the event information are deleted and filtered, and the purity of the event map construction can be effectively improved.
Further, referring to fig. 3, fig. 3 is a flowchart illustrating a step of processing the first event information after being time-stamped in an embodiment of the present application.
Specifically, in step S202, the screening the time-stamped first event information to obtain second event information for constructing an event graph, includes:
step S301, removing sparse events from the first event information subjected to event marking according to a sparse event removal technology, wherein the sparse events are events with incomplete information composition.
Step S302, performing category division on the first event information after the sparse event removal to obtain a plurality of categories.
Step S303, performing duplication elimination processing on the first event information contained in the plurality of categories according to an entity fusion technology to obtain second event information for constructing an event map.
After the event information containing the time information is obtained, event removal processing is carried out on the event information containing the time information, useless event information or event information with incomplete information records in the event information is deleted, useful event information is obtained, and the event information after time marking processing is screened mainly by using a sparse event removal technology to realize event removal.
In the practical application process, when text information is processed by using the TextRank technology to obtain corresponding event information, not all information contained in the event information is complete, and event information with information missing exists, however, the event information has no positive effect on the construction of an event map, and meanwhile, unnecessary operations are generated during map construction, and the purity of the event map can be improved to a certain extent and noise data in the event map can be reduced by filtering the event information with information missing.
When event information is screened, after all event information is obtained based on big data, keywords contained in the event information are obtained, the number of the keywords corresponding to each event information is recorded, event information is classified and counted according to the number of the keywords, the number of events corresponding to the keywords with different numbers is obtained, and finally, partial data are removed according to the fact that the difference of the number of the events is large. For example, when the difference between the number of events corresponding to one keyword and the number of events corresponding to two keywords is the largest, it is determined that the event information corresponding to one keyword is removed as a sparse event.
In some embodiments, after the event information is subjected to the screening processing, the event information obtained by the processing is subjected to the deduplication processing according to an entity fusion technology, where the entity fusion technology is mainly applied to a way of entity fusion normalization, and the event information with the same events recorded in the event information is subjected to the summary normalization processing, that is, the event information with the same events recorded is subjected to the deduplication processing, so that the existence of multiple pieces of event information with the same events recorded in the event library is avoided.
When the duplicate removal processing is carried out, event information can be judged by utilizing a pre-trained logistic regression model, whether repeated event information exists among the event information or not is determined, and the repeated event information is deleted, so that the duplicate removal effect is achieved.
When the event information is subjected to the duplicate removal processing, the process of carrying out the duplicate removal by utilizing entity fusion normalization comprises data grouping, data preprocessing, attribute similarity and entity similarity.
Specifically, data packet: in order to find out all the same entities, if data grouping is not performed, the calculation amount at the time is pairwise comparison, and the calculation amount is too large for massive data, so that the grouping is performed in advance. For each category of data, entity fusion normalization is performed. When event information is grouped, event information with the same keywords corresponding to the event information is divided into the same group or category, after grouping or classification is completed, duplicate removal processing is performed on each group of event information, and repeated event information in each group is filtered and deleted to obtain event information with a single event.
Data preprocessing: the input original data source often has dirty data and data with inconsistent format, the preprocessing and the regularization are time-consuming but have great effect in the actual engineering, and the subsequent algorithm effect is often not good without good data processing. For example, like the hospital name, some sites are directly the xx hospital, some sites are followed by the "xx hospital second class a, the" xx hospital medical insurance fixed-point hospital ", and the like, after taking the data, the name is firstly processed, and the names of the various sites are unified as the" xx hospital ", because the names similar to the" second class a, and the like are used as an attribute value of the hospital.
When the event information is preprocessed, keywords corresponding to the obtained event information are unified, when the keywords are obtained, the keywords and a preset list are inquired and unified, main keywords corresponding to the keywords are determined, and the obtained main keywords are used for replacing the original obtained keywords. For example, the main keyword corresponding to "xx hospital", "xx hospital grade a, etc." and "xx hospital medical insurance fixed point hospital" described above is "xx hospital".
Before calculating the entity similarity, judging the attribute similarity, wherein the attribute similarity comprises the following steps: of the pure string: calculating an editing distance, levenshtein distance, and calculating the distance of the character string A transformed to the character string B through insertion/deletion/replacement operation; set type: calculating the similarity of the Jaccard, and calculating the number of intersection sets/union set sets of the sets; document type: and finding out the keywords of each document through tf-idf, calculating the similarity of the keyword set through cosine similarity, specifically obtaining the keywords in the event information, and then calculating the cosine similarity to obtain the corresponding similarity.
Entity similarity: with the similarity of each attribute of the entity, the similarity of the entity can be calculated. Common methods are divided into two categories: and (3) regression: and directly judging the similarity of the entities according to the similarity of each attribute of the entities. The weight of each attribute similarity can be directly calculated, and the weight of each attribute similarity can also be calculated in a logistic regression mode. Clustering: and directly calculating similar entities through clustering operation. Hierarchical clustering, relevance clustering, Canopy + K-means clustering, and the like can be performed.
Referring to fig. 4, fig. 4 is a schematic block diagram of an early warning event map constructing apparatus according to an embodiment of the present application, which is configured to execute the aforementioned early warning event map constructing method.
As shown in fig. 4, the early warning event map constructing apparatus 400 includes:
a text acquisition module 401, configured to acquire text information for constructing an event map when an event map construction instruction is received;
a first processing module 402, configured to perform event acquisition on the text information, so as to obtain at least one first event information corresponding to the text information and an early warning tag corresponding to the at least one first event information;
a second processing module 403, configured to filter the at least one first event information to obtain at least one second event information for constructing an event graph;
a relationship determining module 404, configured to determine, according to feature information included in the at least one second event, a hierarchical relationship between the at least one second event information;
and the map building module 405 is configured to store the at least one second event information in the corresponding map database according to the hierarchical relationship, so as to complete the building of the event map.
Further, in an embodiment, the text obtaining module 401 is further specifically configured to:
when an event map construction instruction is received, receiving uploaded text information for constructing an event map; or when an event map building instruction is received, reading the input text link, and acquiring text information for building the event map according to the text link.
Further, in an embodiment, the first processing module 402 is further specifically configured to:
performing word segmentation processing on the text information to obtain a plurality of word units; and constructing a graph model, and processing the word units based on the graph model to obtain at least one piece of first event information corresponding to the text information and an early warning label corresponding to the at least one piece of first event information.
Further, in an embodiment, the first processing module 402 is further specifically configured to:
constructing a graph model; the word units are used as the input of the graph model, and the weight values corresponding to the word units in the word units are obtained; and obtaining word units for forming event information according to the weight values so as to obtain at least one first event information corresponding to the text information and an early warning label corresponding to the at least one first event information.
Further, in an embodiment, the second processing module 403 is further specifically configured to:
acquiring time information corresponding to the first event information so as to time-stamp the first event information by using the time information; and screening the first event information subjected to the time marking to obtain second event information subjected to the event map construction.
Further, in an embodiment, the second processing module 403 is further specifically configured to:
removing the sparse events of the time-marked first event information according to a sparse event removal technology, wherein the sparse events are events with incomplete information compositions; classifying the first event information after the sparse event removal to obtain a plurality of classes; and performing duplicate removal processing on the first event information contained in the plurality of categories according to an entity fusion technology to obtain second event information for constructing an event map.
Further, in an embodiment, the map building module 405 is further specifically configured to:
acquiring a hierarchy label corresponding to the second event information, and determining a storage hierarchy of the second event according to the hierarchy label; and storing the second event information according to the storage hierarchy, and establishing an incidence relation between the second event information to complete the construction of the event map.
It should be noted that, as will be clear to those skilled in the art, for convenience and brevity of description, the specific working processes of the apparatus and the modules described above may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.
The apparatus described above may be implemented in the form of a computer program which is executable on a computer device as shown in fig. 5.
Referring to fig. 5, fig. 5 is a schematic block diagram of a computer device according to an embodiment. The computer device may be a server.
Referring to fig. 5, the computer device includes a processor, a memory, and a network interface connected through a system bus, wherein the memory may include a nonvolatile storage medium and an internal memory.
The non-volatile storage medium may store an operating system and a computer program. The computer program includes program instructions that, when executed, cause a processor to perform any one of the methods for risk early warning event graph construction.
The processor is used for providing calculation and control capability and supporting the operation of the whole computer equipment.
The internal memory provides an environment for running a computer program in the non-volatile storage medium, and the computer program, when executed by the processor, can cause the processor to execute any one of the methods for constructing the warning event map.
The network interface is used for network communication, such as sending assigned tasks and the like. Those skilled in the art will appreciate that the architecture shown in fig. 5 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
It should be understood that the Processor may be a Central Processing Unit (CPU), and the Processor may be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, etc. Wherein a general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
Wherein, in one embodiment, the processor is configured to execute a computer program stored in the memory to implement the steps of:
when an event map construction instruction is received, acquiring text information for constructing the event map; event acquisition is carried out on the text information, and at least one piece of first event information corresponding to the text information and an early warning label corresponding to the at least one piece of first event information are obtained; screening the at least one first event information to obtain at least one second event information for constructing an event map; determining a hierarchical relationship between the at least one second event information according to the feature information contained in the at least one second event; and storing the at least one piece of second event information in a corresponding graph database according to the hierarchical relationship so as to complete the construction of the event graph.
In one embodiment, the processor is implementing the method and is further configured to implement:
when an event map construction instruction is received, receiving uploaded text information for constructing an event map; or when an event map building instruction is received, reading the input text link, and acquiring text information for building the event map according to the text link.
In one embodiment, the processor is implementing the method and is further configured to implement:
performing word segmentation processing on the text information to obtain a plurality of word units; and constructing a graph model, and processing the word units based on the graph model to obtain at least one piece of first event information corresponding to the text information and an early warning label corresponding to the at least one piece of first event information.
In one embodiment, the processor is implementing the method and is further configured to implement:
constructing a graph model; the word units are used as the input of the graph model, and the weight values corresponding to the word units in the word units are obtained; and obtaining word units for forming event information according to the weight values so as to obtain at least one first event information corresponding to the text information and an early warning label corresponding to the at least one first event information.
In one embodiment, the processor is implementing the method and is further configured to implement:
acquiring time information corresponding to the first event information so as to time-stamp the first event information by using the time information; and screening the first event information subjected to the time marking to obtain second event information subjected to the event map construction.
In one embodiment, the processor is implementing the method and is further configured to implement:
removing the sparse events of the time-marked first event information according to a sparse event removal technology, wherein the sparse events are events with incomplete information compositions; classifying the first event information after the sparse event removal to obtain a plurality of classes; and performing duplicate removal processing on the first event information contained in the plurality of categories according to an entity fusion technology to obtain second event information for constructing an event map.
In one embodiment, the processor is implementing the method and is further configured to implement:
acquiring a hierarchy label corresponding to the second event information, and determining a storage hierarchy of the second event according to the hierarchy label; and storing the second event information according to the storage hierarchy, and establishing an incidence relation between the second event information to complete the construction of the event map.
The embodiment of the application further provides a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, the computer program comprises program instructions, and the processor executes the program instructions to implement the construction method of any one of the early warning event maps provided by the embodiment of the application.
The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiment, for example, a hard disk or a memory of the computer device. The computer readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like provided on the computer device.
While the invention has been described with reference to specific embodiments, the scope of the invention is not limited thereto, and those skilled in the art can easily conceive various equivalent modifications or substitutions within the technical scope of the invention. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. A construction method of an early warning event map is characterized by comprising the following steps:
when an event map construction instruction is received, acquiring text information for constructing the event map;
event acquisition is carried out on the text information, and at least one piece of first event information corresponding to the text information and an early warning label corresponding to the at least one piece of first event information are obtained;
screening the at least one first event information to obtain at least one second event information for constructing an event map;
determining a hierarchical relationship between the at least one second event information according to the feature information contained in the at least one second event;
and storing the at least one piece of second event information in a corresponding graph database according to the hierarchical relationship so as to complete the construction of the event graph.
2. The method according to claim 1, wherein when receiving the event map construction instruction, acquiring the text information for performing the event map construction comprises:
when an event map construction instruction is received, receiving uploaded text information for constructing an event map; or the like, or, alternatively,
and when an event map building instruction is received, reading the input text link so as to obtain text information for building the event map according to the text link.
3. The method according to claim 1, wherein the obtaining of the event from the text information to obtain at least one first event information corresponding to the text information and the warning label corresponding to the at least one first event information comprises:
performing word segmentation processing on the text information to obtain a plurality of word units;
and constructing a graph model, and processing the word units based on the graph model to obtain at least one piece of first event information corresponding to the text information and an early warning label corresponding to the at least one piece of first event information.
4. The method according to claim 3, wherein the constructing a graph model and processing the word units based on the graph model to obtain at least one first event information corresponding to the text information and an early warning label corresponding to the at least one first event information comprises:
constructing a graph model;
the word units are used as the input of the graph model, and the weight values corresponding to the word units in the word units are obtained;
and obtaining word units for forming event information according to the weight values so as to obtain at least one first event information corresponding to the text information and an early warning label corresponding to the at least one first event information.
5. The method according to claim 3, wherein the filtering the at least one first text message to obtain at least one second event message for event graph construction comprises:
acquiring time information corresponding to the first event information so as to time-stamp the first event information by using the time information;
and screening the first event information subjected to the time marking to obtain second event information subjected to the event map construction.
6. The method of claim 5, wherein the filtering the time-stamped first event information to obtain second event information for event graph construction comprises:
removing the sparse events of the time-marked first event information according to a sparse event removal technology, wherein the sparse events are events with incomplete information compositions;
classifying the first event information after the sparse event removal to obtain a plurality of classes;
and performing duplicate removal processing on the first event information contained in the plurality of categories according to an entity fusion technology to obtain second event information for constructing an event map.
7. The method according to claim 5 or 6, wherein said storing said second event information in a corresponding graph database to complete the construction of an event graph comprises:
acquiring a hierarchy label corresponding to the second event information, and determining a storage hierarchy of the second event according to the hierarchy label;
and storing the second event information according to the storage hierarchy, and establishing an incidence relation between the second event information to complete the construction of the event map.
8. An early warning event map construction device, which is characterized by comprising:
the text acquisition module is used for acquiring text information for constructing the event map when receiving an event map construction instruction;
the first processing module is used for performing event acquisition on the text information to obtain at least one piece of first event information corresponding to the text information and an early warning label corresponding to the at least one piece of first event information;
the second processing module is used for screening the at least one first event information to obtain at least one second event information for constructing an event map;
the relationship determination module is used for determining the hierarchical relationship among the at least one piece of second event information according to the characteristic information contained in the at least one piece of second event information;
and the map construction module is used for storing the at least one piece of second event information in a corresponding map database according to the hierarchical relationship so as to complete the construction of the event map.
9. A computer device comprising a memory and a processor, the memory having stored therein computer readable instructions which, when executed by the processor, cause the processor to perform the steps of the warning event map construction method of any one of claims 1 to 7.
10. A computer readable storage medium storing a computer program, wherein the computer readable instructions, when executed by the processor, cause one or more processors to perform the steps of the method of constructing a warning event map according to any one of claims 1 to 7.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010136834.0A CN111475612A (en) | 2020-03-02 | 2020-03-02 | Construction method, device and equipment of early warning event map and storage medium |
| PCT/CN2021/070933 WO2021175009A1 (en) | 2020-03-02 | 2021-01-08 | Early warning event graph construction method and apparatus, device, and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010136834.0A CN111475612A (en) | 2020-03-02 | 2020-03-02 | Construction method, device and equipment of early warning event map and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111475612A true CN111475612A (en) | 2020-07-31 |
Family
ID=71747120
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010136834.0A Pending CN111475612A (en) | 2020-03-02 | 2020-03-02 | Construction method, device and equipment of early warning event map and storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN111475612A (en) |
| WO (1) | WO2021175009A1 (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112749239A (en) * | 2021-01-20 | 2021-05-04 | 青岛海信网络科技股份有限公司 | Event map construction method and device and computing equipment |
| WO2021175009A1 (en) * | 2020-03-02 | 2021-09-10 | 深圳壹账通智能科技有限公司 | Early warning event graph construction method and apparatus, device, and storage medium |
| CN113449116A (en) * | 2021-06-22 | 2021-09-28 | 青岛海信网络科技股份有限公司 | Map construction and early warning method, device and medium |
| 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 |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113392220B (en) * | 2020-10-23 | 2024-03-26 | 腾讯科技（深圳）有限公司 | Knowledge graph generation method and device, computer equipment and storage medium |
| CN114647743B (en) * | 2022-05-20 | 2022-08-26 | 国网浙江省电力有限公司 | Method and device for generating and processing power marketing full-service access control rule map |
| CN114707004B (en) * | 2022-05-24 | 2022-08-16 | 国网浙江省电力有限公司信息通信分公司 | Method and system for extracting and processing case-affair relation based on image model and language model |
| CN115146081B (en) * | 2022-08-31 | 2022-12-09 | 合肥中科迪宏自动化有限公司 | Construction method and diagnosis method of fault diagnosis knowledge graph of production equipment |
| CN115630846B (en) * | 2022-12-07 | 2023-08-01 | 速度科技股份有限公司 | Processing method suitable for natural disaster risk comprehensive monitoring data |
| CN116054910A (en) * | 2022-12-20 | 2023-05-02 | 中国人民解放军63819部队 | Fault analysis and device for earth station equipment based on knowledge graph construction |
| CN116842200A (en) * | 2023-03-29 | 2023-10-03 | 全景智联(武汉)科技有限公司 | Event file aggregation management method |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107783973B (en) * | 2016-08-24 | 2022-02-25 | 慧科讯业有限公司 | Method, device and system for monitoring internet media event based on industry knowledge map database |
| WO2018209254A1 (en) * | 2017-05-11 | 2018-11-15 | Hubspot, Inc. | Methods and systems for automated generation of personalized messages |
| CN108763333B (en) * | 2018-05-11 | 2022-05-17 | 北京航空航天大学 | Social media-based event map construction method |
| CN109977237B (en) * | 2019-05-27 | 2019-09-10 | 南京擎盾信息科技有限公司 | A kind of dynamic law occurrence diagram spectrum construction method towards legal field |
| CN110781317B (en) * | 2019-10-29 | 2022-03-01 | 北京明略软件系统有限公司 | Method and device for constructing event map and electronic equipment |
| CN111475612A (en) * | 2020-03-02 | 2020-07-31 | 深圳壹账通智能科技有限公司 | Construction method, device and equipment of early warning event map and storage medium |
-
2020
- 2020-03-02 CN CN202010136834.0A patent/CN111475612A/en active Pending
-
2021
- 2021-01-08 WO PCT/CN2021/070933 patent/WO2021175009A1/en active Application Filing
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021175009A1 (en) * | 2020-03-02 | 2021-09-10 | 深圳壹账通智能科技有限公司 | Early warning event graph construction method and apparatus, device, and storage medium |
| CN112749239A (en) * | 2021-01-20 | 2021-05-04 | 青岛海信网络科技股份有限公司 | Event map construction method and device and computing equipment |
| CN113449116A (en) * | 2021-06-22 | 2021-09-28 | 青岛海信网络科技股份有限公司 | Map construction and early warning method, device and medium |
| 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 |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2021175009A1 (en) | 2021-09-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111475612A (en) | Construction method, device and equipment of early warning event map and storage medium | |
| US10733193B2 (en) | Similar document identification using artificial intelligence | |
| US8630989B2 (en) | Systems and methods for information extraction using contextual pattern discovery | |
| CN110851598B (en) | Text classification method and device, terminal equipment and storage medium | |
| JP5534280B2 (en) | Text clustering apparatus, text clustering method, and program | |
| US20130097168A1 (en) | Method to identify common structures in formatted text documents | |
| CN108021651B (en) | Network public opinion risk assessment method and device | |
| CN111259160B (en) | Knowledge graph construction method, device, equipment and storage medium | |
| US9270749B2 (en) | Leveraging social media to assist in troubleshooting | |
| CN111460162B (en) | Text classification method and device, terminal equipment and computer readable storage medium | |
| CN113590824A (en) | Method and device for constructing causal graph and related equipment | |
| CN112328857B (en) | Product knowledge aggregation method and device, computer equipment and storage medium | |
| CN113723737A (en) | Enterprise portrait-based policy matching method, device, equipment and medium | |
| WO2023071127A1 (en) | Policy recommended method and apparatus, device, and storage medium | |
| CN110737770B (en) | Text data sensitivity identification method and device, electronic equipment and storage medium | |
| CN107315807B (en) | Talent recommendation method and device | |
| CN116260866A (en) | Government information pushing method and device based on machine learning and computer equipment | |
| CN113705201B (en) | Text-based event probability prediction evaluation algorithm, electronic device and storage medium | |
| CN115098585A (en) | Automatic law and regulation data processing method and system based on big data | |
| CN115858805A (en) | Knowledge graph construction management system and method | |
| US10776399B1 (en) | Document classification prediction and content analytics using artificial intelligence | |
| CN114281998A (en) | Multi-level annotator-oriented event annotation system construction method based on crowdsourcing technology | |
| JP2010170287A (en) | Data extraction system | |
| CN114003787A (en) | Data visualization method based on artificial intelligence and related equipment | |
| CN113361585A (en) | Method for optimizing and screening clues based on supervised learning algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 40033036 Country of ref document: HK |
|
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |