CN114706992A - Event information processing system based on knowledge graph - Google Patents
Event information processing system based on knowledge graph Download PDFInfo
- Publication number
- CN114706992A CN114706992A CN202210147145.9A CN202210147145A CN114706992A CN 114706992 A CN114706992 A CN 114706992A CN 202210147145 A CN202210147145 A CN 202210147145A CN 114706992 A CN114706992 A CN 114706992A
- Authority
- CN
- China
- Prior art keywords
- event
- sample
- entity
- database
- obtaining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—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
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
Abstract
The invention provides a knowledge graph-based event information processing system, which comprises: a first database, a second database, a processor and a memory storing a computer program, the first database comprising: a sample event ID and a sample text list corresponding to the sample event ID, which when executed by a processor, implement the steps of: acquiring a sample text corresponding to the sample event ID from a first database, and acquiring a multi-component list corresponding to the sample text according to the sample text; acquiring training set data according to all sample texts and the multi-element list corresponding to the sample texts; and inputting the acquired training set data into the multivariate construction model for training to obtain the target multivariate construction model. The method and the device can construct a training set according to actual sample events and acquire the target multi-component set construction model, so that different descriptions of the same entity can be accurately identified, and the accuracy and the applicability of the model are improved.
Description
Technical Field
The invention relates to the field of data processing, in particular to an event information processing system based on a knowledge graph.
Background
With the rapid popularization and development of the internet, a great deal of data information is generated and spread in the network, and how to timely and accurately find needed information from a great amount of natural language texts becomes increasingly urgent. The massive natural language documents have the characteristics of large data volume, non-uniform structure, high redundancy, quick update and the like. In the prior art, an event extraction model is usually obtained by training in a machine learning manner to extract events that are of interest to a user from unstructured information, and the events are presented to the user in a structured manner. However, the method of directly extracting events by using an event extraction model depends on keywords, and if the number of the keywords is small, incomplete or inappropriate, the method has a great influence on the event extraction result, and particularly for the event types which are not used as training samples and are subjected to learning, the accuracy of event extraction is low, and the extracted event information is incomplete. Therefore, how to improve the integrity and accuracy of the event extraction result is a technical problem to be solved urgently.
Disclosure of Invention
Aiming at the technical problems, the technical scheme adopted by the invention is as follows:
a knowledge-graph based event information processing system, the system comprising: a first database, a second database, a processor and a memory storing a computer program, the first database comprising: the second database comprises a sample event ID and a sample text list corresponding to the sample event ID: the computer program, when executed by a processor, implements the following steps:
s100, obtaining A = (A) from the database1,A2,……,Am),Ai=(Ai1,Ai2,……, ) Wherein A isijI =1 … … m, where m is the number of sample events, and j =1 … … ni,niThe number of all sample texts in a sample text list corresponding to the ith sample event ID is determined;
s200, according to AijObtaining AijCorresponding initial entity list (A)1 ij、A2 ij,……,Ap ij) Wherein A isq ijMeans AijThe corresponding qth initial entity, q =1 … … p, p being the number of initial entities;
s300, according to Aq ijObtaining AiIntermediate data set of corresponding sample event ID=(A1 i,A2 i,……,Ap i) Wherein A isq i=(Aq i1、Aq i2、……,);
s500, based on all AiAnd constructing a training set of the corresponding sample event ID into target training set data.
S600, inputting the target training set data into the event graph model for training to obtain a target event graph model.
The invention provides a knowledge graph-based event information processing system, which comprises: a first database, a second database, a processor and a memory storing a computer program, the first database comprising: a sample event ID and a sample text list corresponding to the sample event ID, which when executed by a processor, performs the steps of: acquiring a sample text corresponding to the sample event ID from the first database, and acquiring a multi-component list corresponding to the sample text according to the sample text; acquiring training set data according to all sample texts and the multi-element list corresponding to the sample texts; the acquired training set data is input into the multivariate construction model for training to obtain the target multivariate group construction model, the training set can be constructed according to the actual sample event, and the target multivariate group construction model is acquired, so that the training accuracy and the applicability of the model in practical application are improved.
In addition, the system acquires a target entity list corresponding to the original event ID; comparing any target entity in a target entity list corresponding to the original event ID with a preset threshold region corresponding to the target entity to obtain the priority corresponding to the target entity; acquiring the priority corresponding to the original event ID and the report times corresponding to the original event ID based on the priority corresponding to the target entity; determining the actual priority corresponding to the original event ID according to the priority corresponding to the original event ID, the reporting times corresponding to the original event ID and the preset reporting time conditions; when the priority corresponding to the original event ID is equal to the actual priority corresponding to the original event ID, adjusting all initial weights in the priority corresponding to the original event ID to obtain a target weight corresponding to the initial weights; acquiring an original event ID, and acquiring a priority corresponding to the original event ID according to a target entity list of the original event ID and all adjusted target weights, so that an event message of the original event ID is sent according to the priority corresponding to the original event ID; the method and the device can avoid the situation that the user cannot know the important event at the first time due to the fact that the event message is sent in a delayed mode.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, 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 flow diagram of a knowledge-graph based event information processing system according to an embodiment of the present invention;
FIG. 2 is a flow diagram of another knowledge-graph based event information processing system according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. 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 invention.
As shown in fig. 1, an embodiment of the present invention provides a knowledge-graph-based event information processing system, which includes: a database, a processor, and a memory storing a computer program, the database comprising: a sample event ID and a sample text list corresponding to the sample event ID, which when executed by a processor, performs the steps of:
s100, obtaining A = (A) from the database1,A2,……,Am),Ai=(Ai1,Ai2,……,) Wherein A isijI =1 … … m, where m is the number of sample events, and j =1 … … ni,niThe number of all sample texts in the sample text list corresponding to the ith sample event ID is obtained.
Specifically, the sample event ID refers to a unique identifier that characterizes the identity of the sample event; the sample event corresponding to the sample event ID is an event occurring within a preset time period.
Further, the value range of the preset time period is 1-3 years, and preferably, the value of the preset time period is 3 years.
Specifically, the sample text refers to text crawled from an information platform and used for describing sample events.
S200, according to AijObtaining AijCorresponding initial entity list (A)1 ij、A2 ij,……,Ap ij) Wherein A isq ijMeans AijThe corresponding qth initial entity, q =1 … … p, p is the initial entity number.
Specifically, the database in the system further includes: the method comprises the steps of obtaining a sample event ID, an event type corresponding to the sample event ID, a preset multi-element list corresponding to the event type and a preset trigger word list corresponding to each preset multi-element group.
Further, those skilled in the art know that in the step S200, the following steps are also included:
s201, obtaining AijAccording to said trigger word AijThe trigger word is compared with each preset trigger word in the preset trigger word list, and those skilled in the art know the method for obtaining the text trigger word, which is not described herein again.
S203, place AijWhen the trigger word is consistent with any preset trigger word in the preset trigger word database, determining AijCan be understood as: when A is inijWhen the trigger word is consistent with any preset trigger word in the preset trigger word list, obtaining AijCorresponding sample file ID, according to AijCorresponding sample file ID, get AijThe type of event(s).
S205, obtaining A from the second databaseijAccording to the event type of (A)ijObtaining a preset tuple corresponding to the event type of AijA corresponding initial entity list; it can be understood that: according to AijCorresponding initial entity list and AijThe event types of (2) are consistent with the corresponding preset multi-component groups.
Preferably, when p =3, aijCorresponding multigroup list (A)1 ij、A2 ij,A3 ij) Wherein A is1 ijFor the ith sampleThe first initial entity in the jth sample text in the piece, A2 ijIs a second initial entity in the jth sample text in the ith sample event, A3 ijMeans A1 ijAnd A2 ijAnd as a third entity; it can be understood that: when A isijThe corresponding event type is a natural disaster event, e.g. A1 ijAs a seismic source address, A2 ijIs time, A3 ijIs the occurrence of a 2.0 level earthquake.
S300, according to Aq ijObtaining AiIntermediate data set of corresponding sample event ID=(A1 i,A2 i,……,Ap i) Wherein A isq i=(Aq i1、Aq i2、……,) (ii) a It can be understood that: and constructing an entity list by using single entities corresponding to different sample texts of the same sample event ID.
Specifically, the step S400 further includes the steps of:
s401, traverse Aq iObtaining Aq iCorresponding entity number list Bq i=(Bq i1,Bq i2,……,),Bq ixIs referred to as Aq iNumber of xth entity class, where x =1 … … sq,sqIs at Aq iThe number of kinds of the qth entity in (1).
S403, according to Bq ixObtaining Bq ixCorresponding probability value Fq ix,Fq ixThe following conditions are met: fq ix=Bq ix/Bq i0Wherein B isq i0Means Bq iThe median maximum magnitude value.
S405, when Fq ix> predetermined probability threshold F0When determining Fq ixCorresponding entity as intermediate entity, constructing Fq ixCorresponding intermediate entity list and determine Bq i0Corresponding entity as key entity Hq i0。
Specifically, F0The value range of (A) is 0.1-0.3; preferably, F0The value range of (a) is 0.2.
S407, traverse Fq ixCorresponding intermediate entity list and from Fq ixObtaining H from corresponding intermediate entity listq i0Corresponding all associated entities (H)q i1,Hq i2,……,Hq ikq),kqIs Hq i0The corresponding number of associated entities.
Specifically, the associated entity corresponding to the key entity refers to other intermediate entities in the intermediate entity list except the key entity; it can be understood that: the associated entity of the key entity characterizes the same meaning as the key entity. By taking the key entities and the associated entities of the key entities as training sets, synonyms or near synonyms in the text can be identified, so that the accuracy of model identification is improved, and the complexity of data processing is reduced.
S409, adding Hq i0And Hq i0Corresponding all associated entities (H)q i1,Hq i2,……,Hq ik) Constructed ofAq iCorresponding key entity list Hq i=(Hq i0,Hq i1,Hq i2,……,Hq ikq) And based on Hq iIs constructed asiA training set of corresponding sample event IDs.
In particular, the amount of the solvent to be used,removing A inq iThe list of key entities corresponding to other initial entities except the corresponding initial entity can refer to Aq iA corresponding list of key entities is determined.
Specifically, in step S409, a is addediThe key entity list corresponding to all the initial entities is constructed as AiA training set of corresponding sample event IDs.
S500, based on all AiAnd constructing a training set of the corresponding sample event ID into target training set data.
S600, inputting target training set data into an event graph model for training to obtain a target event graph model; those skilled in the art will appreciate that any event graph model known in the art may be used and will not be described in detail herein; the method can accurately and quickly extract the entities with different meanings of the same text in the event map model, and avoids the problem that the event map cannot be established due to the fact that the entities with the same meaning are omitted.
In a specific embodiment, the computer program, when executed by a processor, in the system further implements the following steps, as shown in fig. 2:
s1, acquiring a target entity list D = (D) corresponding to the original event ID1,D2,……,Dg),DyRefers to the y-th target entity of the original event ID, y =1 … … g, g is the number of target entities of the original event ID.
Specifically, the step S1 further includes the following step D:
and S11, acquiring all target texts corresponding to the original event IDs.
Specifically, any target text corresponding to the original event ID is consistent with the obtaining manner of the sample text in the above example, and is not described herein again.
And S12, inputting all target texts corresponding to the original event IDs into the target event graph model, wherein all key entity lists corresponding to the original event IDs are included.
Specifically, any key entity list corresponding to the original event ID is consistent with the key entity list corresponding to the sample event ID in the above example, and details are not repeated here.
And S13, traversing the key entity list corresponding to any original event ID, and taking the key entity corresponding to the maximum probability value in the key entity list corresponding to the original event ID as a target entity.
Specifically, the probability value of the key entity in the key entity list corresponding to the original event ID is consistent with the obtaining manner of the initial entity probability value corresponding to the sample event ID in the above example, which is not described herein again.
S2, mixing DyAnd DyComparing the corresponding preset threshold value areas to obtain DyCorresponding priority Cy。
In a specific embodiment, D is also determined in step S2 byyThe corresponding preset threshold region:
and S21, acquiring the event type corresponding to the original event ID as a preset event type.
S22, acquiring all sample event IDs corresponding to the preset event types from A as initial event IDs, and constructing an initial event ID list U = (U)1,U2,……,Uf) Wherein, UtRefers to the t-th initial event ID, t =1 … … f, f is the number of initial event IDs.
S23, acquiring U from AtConstructing U from all corresponding sample textstCorresponding target entity list Qt=(Qt1,Qt2,……,Qtg) Wherein Q istyRefers to UtG is UtThe number of target entities.
Specifically, the target entity list of the initial event ID in the step S23 can be obtained by referring to the step S1, which is not described herein again.
S24, obtaining an intermediate entity list Q'y=(Q1y,Q2y……,Qfy)。
S25, according to Q'yObtaining Py,PyThe following conditions are met:
further, RyThe following conditions are met:
further, EyThe following conditions are met:
s25, according to PyIs divided into z DyA corresponding preset threshold region; it can be understood that: with PyIs a value range and is represented by RyFor the central point, a person skilled in the art divides z value ranges according to actual requirements to serve as a preset threshold region, and each preset threshold region corresponds to a preset priority, for example, the magnitude is divided into (0, 1), [1, 3), [3, 4.5), [4.5, 6), [6, 7), [7, 8), [8, + ∞), and when an event related to the magnitude of 5 is required to be reported, only the event is acquired in a text corresponding to [4.5, 6), so that the efficiency of event extraction is improved.
S3, based on CyAcquiring the priority C corresponding to the original event ID, wherein the priority C meets the following conditions:
And S4, acquiring the report times T corresponding to the original event ID.
S5, determining the actual priority C0 corresponding to the original event ID according to the preset conditions of T and the reporting times.
Specifically, the report frequency preset condition may determine the divided regions and the corresponding priorities of the regions according to a normal distribution method.
S6, when C ≠ C0When, adjust all W in CyTo obtain WyCorresponding target weight(ii) a Those skilled in the art will appreciate that the adjustment method may be any one of the prior art, and may be selected according to actual conditions.
By adjusting the attribute weights of the events according to the priorities in the actual sample text, the accuracy of the weight calculation process can be improved.
In a specific embodiment, further comprising the step of determining Wy:
According to DyObtaining the current time period D from the second databaseyCorresponding weight list W'y=(W'y1,W'y2,……,W'yβ) And D within a preset time periodyCorresponding weight value epsilon, wherein W'yαIs referred to as DyThe weight value of the corresponding alpha-th preset tuple list, alpha =1 … … beta;
preferably, the current time period is from the same date of the previous year to the current date; the preset time period is from the current date of the previous two years to the current date of the previous year, for example, the current date is 12/21/2021 year, the current time period is from 12/21/2020/12/2021 year to 21/12/2021 year, and the preset time period is from 12/21/2019 year to 21/12/2020/12/21 year.
Obtaining D from third party weight platformyA base value η of each keyword in (1) in several aspects;
according to W' y, determining that Wy meets the following conditions:
s7, obtaining a target event ID, obtaining a target entity list according to the target event ID and all W 'after adjustment'yObtaining the priority corresponding to the target event ID, so as to send the event message of the target event ID according to the priority corresponding to the target event ID; the method and the device can avoid the situation that the user cannot know the important event at the first time due to the fact that the event message is sent in a delayed mode.
Specifically, the event type corresponding to the target event ID is consistent with the event type corresponding to the original event ID.
The invention provides a knowledge graph-based event information processing system, which comprises: a first database, a second database, a processor and a memory storing a computer program, the first database comprising: the second database comprises a sample event ID and a sample text list corresponding to the sample event ID: the computer program, when executed by a processor, implements the following steps: acquiring a sample text corresponding to the sample event ID from the first database, and acquiring a multi-component list corresponding to the sample text according to the sample text; acquiring training set data according to all sample texts and the multi-element list corresponding to the sample texts; and inputting the acquired training set data into the multivariate construction model for training to obtain the target multivariate construction model. The training set is constructed according to the actual sample event, and the target multi-component set construction model is obtained, so that the training accuracy and the applicability of the model in practical application are improved. The target events are weight-adjusted by building a model using the target multivariate set.
In addition, the system acquires a target entity list corresponding to the original event ID; comparing any target entity in a target entity list corresponding to the original event ID with a preset threshold region corresponding to the target entity to obtain the priority corresponding to the target entity; acquiring the priority corresponding to the original event ID and the report times corresponding to the original event ID based on the priority corresponding to the target entity; determining the actual priority corresponding to the original event ID according to the priority corresponding to the original event ID, the reporting times corresponding to the original event ID and the preset conditions of the reporting times; when the priority corresponding to the original event ID is equal to the actual priority corresponding to the original event ID, adjusting all initial weights in the priority corresponding to the original event ID to obtain a target weight corresponding to the initial weights; acquiring an original event ID, and acquiring a priority corresponding to the original event ID according to a target entity list of the original event ID and all adjusted target weights, so that an event message of the original event ID is sent according to the priority corresponding to the original event ID; the method and the device can avoid the situation that the user cannot know the important event at the first time due to the fact that the event message is sent in a delayed mode.
Although some specific embodiments of the present invention have been described in detail by way of illustration, it should be understood by those skilled in the art that the above illustration is only for the purpose of illustration and is not intended to limit the scope of the invention. It will also be appreciated by those skilled in the art that various modifications may be made to the embodiments without departing from the scope and spirit of the invention. The scope of the invention is defined by the appended claims.
Claims (10)
1. A knowledge-graph based event information processing system, the system comprising: a first database, a second database, a processor and a memory storing a computer program, the first database comprising: a sample event ID and a sample text list corresponding to the sample event ID, which when executed by a processor, performs the steps of:
s100, obtaining A = (A) from the database1,A2,……,Am),Ai=(Ai1,Ai2,……,) Wherein A isijI =1 … … m, where m is the number of sample events, and j =1 … … ni,niFor the ith sample eventThe number of all sample texts in a sample text list corresponding to the ID;
s200, according to AijObtaining AijCorresponding initial entity list (A)1 ij、A2 ij,……,Ap ij) Wherein A isq ijMeans AijThe corresponding qth initial entity, q =1 … … p, p being the number of initial entities;
s300, according to Aq ijObtaining AiIntermediate data set of corresponding sample event ID=(A1 i,A2 i,……,Ap i) Wherein A isq i=(Aq i1、Aq i2、……, );
s500, based on all AiConstructing a training set of the corresponding sample event ID into target training set data;
s600, inputting the target training set data into the event graph model for training to obtain a target event graph model.
2. The knowledge-graph-based event information processing system of claim 1, wherein further comprising a second database in the system comprises: the event type corresponding to the sample event ID and the preset multi-component list corresponding to the event type are obtained.
3. The system of claim 2, further comprising, in step S200, the steps of:
s201, obtaining AijAnd according to AijComparing the trigger word with each preset trigger word in a preset trigger word database;
s203, according to AijWhen the trigger word is consistent with any one of the preset trigger words in the preset trigger word database, B is determinedijThe type of event of (2);
s205, obtaining A from the second databaseijIs corresponding to the event type and is according to AijObtaining a preset multi-component list corresponding to the event type of AijCorresponding list of tuples.
4. The system of claim 1, further comprising, in step S300, the steps of:
S303, traverse Aq iObtaining the q-th entity number list Bq i=(Bq i1,Bq i2,……,),Bq ixRefers to the q-th entity number of the x-th class corresponding to the ith sample event ID, wherein x =1 … … si,siThe category number of the q entity corresponding to the ith sample event ID;
s305, according to Bq i,Bq ixCorresponding first probability value Fq ix,Fq ixThe following conditions are met: fq ix=Bq ix/Bq i0Wherein, Bq i0Is Bq iA medium maximum value;
s307, when Fq ixNot less than the preset probability threshold F0When determining Fq ixThe corresponding qth entity is taken as an intermediate entity;
s309, traverse Bqi, obtaining the maximum entity number Bq ixAnd constructing a q-th sample data list by taking the corresponding entity as a key entity according to the key entity and the associated entity of the key entity.
5. The knowledge-graph-based event information processing system of claim 3, wherein the associated entity of the key entity refers to other intermediate entities than the key entity among all intermediate entities.
6. The system of claim 1, wherein the sample event corresponding to the sample event ID is an event occurring within a preset time period, and wherein the preset time period has a value range of 1 to 3 years.
7. The knowledge-graph based event information processing system of claim 1 wherein all sample data lists are determined in a consistent manner.
8. The knowledge-graph-based event information processing system of claim 1, wherein p = 3.
9. The knowledgegraph-based event information handling system of claim 7, wherein a when p =3ijCorresponding multigroup list (A)1 ij、A2 ij,A3 ij) Wherein A is1 ijIs the first entity in the jth sample text in the ith sample event, A2 ijFor the second entity in the jth sample text in the ith sample event, A3 ijMeans A1 ijAnd A2 ijThe relationship between them.
10. The knowledge-graph based event information processing system of claim 3 wherein all sample data lists are determined in a consistent manner.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210147145.9A CN114706992B (en) | 2022-02-17 | 2022-02-17 | Event information processing system based on knowledge graph |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210147145.9A CN114706992B (en) | 2022-02-17 | 2022-02-17 | Event information processing system based on knowledge graph |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114706992A true CN114706992A (en) | 2022-07-05 |
CN114706992B CN114706992B (en) | 2022-09-30 |
Family
ID=82166779
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210147145.9A Active CN114706992B (en) | 2022-02-17 | 2022-02-17 | Event information processing system based on knowledge graph |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114706992B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116304117A (en) * | 2023-05-15 | 2023-06-23 | 北京睿企信息科技有限公司 | Data processing method, system and storage medium for acquiring text information |
CN117435697A (en) * | 2023-12-21 | 2024-01-23 | 中科雨辰科技有限公司 | Data processing system for acquiring core event |
CN117520754A (en) * | 2024-01-05 | 2024-02-06 | 北京睿企信息科技有限公司 | Pretreatment system for model training data |
Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180069885A1 (en) * | 2016-09-06 | 2018-03-08 | Accenture Global Solutions Limited | Graph database analysis for network anomaly detection systems |
CN108763333A (en) * | 2018-05-11 | 2018-11-06 | 北京航空航天大学 | A kind of event collection of illustrative plates construction method based on Social Media |
CN109670174A (en) * | 2018-12-14 | 2019-04-23 | 腾讯科技(深圳)有限公司 | A kind of training method and device of event recognition model |
CN110765265A (en) * | 2019-09-06 | 2020-02-07 | 平安科技(深圳)有限公司 | Information classification extraction method and device, computer equipment and storage medium |
CN111143576A (en) * | 2019-12-18 | 2020-05-12 | 中科院计算技术研究所大数据研究院 | Event-oriented dynamic knowledge graph construction method and device |
CN111143578A (en) * | 2019-12-30 | 2020-05-12 | 智慧神州(北京)科技有限公司 | Method, device and processor for extracting event relation based on neural network |
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 |
CN111581396A (en) * | 2020-05-06 | 2020-08-25 | 西安交通大学 | Event graph construction system and method based on multi-dimensional feature fusion and dependency syntax |
CN111737552A (en) * | 2020-06-04 | 2020-10-02 | 中国科学院自动化研究所 | Method, device and equipment for extracting training information model and acquiring knowledge graph |
CN112269885A (en) * | 2020-11-16 | 2021-01-26 | 北京百度网讯科技有限公司 | Method, apparatus, device and storage medium for processing data |
CN113157931A (en) * | 2021-01-20 | 2021-07-23 | 青岛海信网络科技股份有限公司 | Fusion map construction method and device |
CN113312500A (en) * | 2021-06-24 | 2021-08-27 | 河海大学 | Method for constructing event map for safe operation of dam |
CN113553442A (en) * | 2021-07-13 | 2021-10-26 | 西安邮电大学 | Unsupervised event knowledge graph construction method and system |
CN113590737A (en) * | 2021-09-28 | 2021-11-02 | 中国人民解放军国防科技大学 | Event data processing method, device, equipment and medium based on knowledge graph |
-
2022
- 2022-02-17 CN CN202210147145.9A patent/CN114706992B/en active Active
Patent Citations (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20180069885A1 (en) * | 2016-09-06 | 2018-03-08 | Accenture Global Solutions Limited | Graph database analysis for network anomaly detection systems |
CN108763333A (en) * | 2018-05-11 | 2018-11-06 | 北京航空航天大学 | A kind of event collection of illustrative plates construction method based on Social Media |
CN109670174A (en) * | 2018-12-14 | 2019-04-23 | 腾讯科技(深圳)有限公司 | A kind of training method and device of event recognition model |
CN110765265A (en) * | 2019-09-06 | 2020-02-07 | 平安科技(深圳)有限公司 | Information classification extraction method and device, computer equipment and storage medium |
CN111143576A (en) * | 2019-12-18 | 2020-05-12 | 中科院计算技术研究所大数据研究院 | Event-oriented dynamic knowledge graph construction method and device |
CN111143578A (en) * | 2019-12-30 | 2020-05-12 | 智慧神州(北京)科技有限公司 | Method, device and processor for extracting event relation based on neural network |
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 |
CN111581396A (en) * | 2020-05-06 | 2020-08-25 | 西安交通大学 | Event graph construction system and method based on multi-dimensional feature fusion and dependency syntax |
CN111737552A (en) * | 2020-06-04 | 2020-10-02 | 中国科学院自动化研究所 | Method, device and equipment for extracting training information model and acquiring knowledge graph |
CN112269885A (en) * | 2020-11-16 | 2021-01-26 | 北京百度网讯科技有限公司 | Method, apparatus, device and storage medium for processing data |
CN113157931A (en) * | 2021-01-20 | 2021-07-23 | 青岛海信网络科技股份有限公司 | Fusion map construction method and device |
CN113312500A (en) * | 2021-06-24 | 2021-08-27 | 河海大学 | Method for constructing event map for safe operation of dam |
CN113553442A (en) * | 2021-07-13 | 2021-10-26 | 西安邮电大学 | Unsupervised event knowledge graph construction method and system |
CN113590737A (en) * | 2021-09-28 | 2021-11-02 | 中国人民解放军国防科技大学 | Event data processing method, device, equipment and medium based on knowledge graph |
Non-Patent Citations (2)
Title |
---|
王毅等: "领域事件图谱构建方法综述", 《数据分析与知识发现》 * |
白璐: "面向政治领域的事理演化图谱构建", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116304117A (en) * | 2023-05-15 | 2023-06-23 | 北京睿企信息科技有限公司 | Data processing method, system and storage medium for acquiring text information |
CN116304117B (en) * | 2023-05-15 | 2023-09-08 | 北京睿企信息科技有限公司 | Data processing method, system and storage medium for acquiring text information |
CN117435697A (en) * | 2023-12-21 | 2024-01-23 | 中科雨辰科技有限公司 | Data processing system for acquiring core event |
CN117435697B (en) * | 2023-12-21 | 2024-03-22 | 中科雨辰科技有限公司 | Data processing system for acquiring core event |
CN117520754A (en) * | 2024-01-05 | 2024-02-06 | 北京睿企信息科技有限公司 | Pretreatment system for model training data |
CN117520754B (en) * | 2024-01-05 | 2024-04-12 | 北京睿企信息科技有限公司 | Pretreatment system for model training data |
Also Published As
Publication number | Publication date |
---|---|
CN114706992B (en) | 2022-09-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114706992B (en) | Event information processing system based on knowledge graph | |
US11190562B2 (en) | Generic event stream processing for machine learning | |
US9053171B2 (en) | Clustering data points | |
CN111698247B (en) | Abnormal account detection method, device, equipment and storage medium | |
US9218568B2 (en) | Disambiguating data using contextual and historical information | |
CN113282960B (en) | Privacy calculation method, device, system and equipment based on federal learning | |
WO2021159738A1 (en) | Data recommendation method and device based on medical field, and server and storage medium | |
KR102104316B1 (en) | Apparatus for predicting stock price of company by analyzing news and operating method thereof | |
CN110709826A (en) | Method and system for linking data records from heterogeneous databases | |
WO2019085463A1 (en) | Department demand recommendation method, application server, and computer-readable storage medium | |
WO2016206557A1 (en) | Risk identification method and apparatus | |
US20230121679A1 (en) | Inference-Based Assignment of Data Type to Data | |
US20240004939A1 (en) | Random sampling from a search engine | |
CN108710664B (en) | Hot word analysis method, computer readable storage medium and terminal device | |
WO2019001463A1 (en) | Data processing method and apparatus | |
CN111355725B (en) | Method and device for detecting network intrusion data | |
CN112948274A (en) | Test case scoring model training method and test case selection method | |
CN114493142A (en) | Method, device, equipment and storage medium for matching support policy with enterprise | |
WO2022217712A1 (en) | Data mining method and apparatus, and computer device and storage medium | |
WO2021129368A1 (en) | Method and apparatus for determining client type | |
JP6785693B2 (en) | Information processing systems, information processing methods, and programs | |
WO2022068659A1 (en) | Information pushing method and apparatus and storage medium | |
Wang et al. | Intelligent weight generation algorithm based on binary isolation tree | |
US20230418825A1 (en) | Database system, database structure, and associated methods | |
Kajimura et al. | Quality control for crowdsourced POI collection |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |