CN112199583B - Network public opinion information intelligent processing method and system based on multi-rule association analysis - Google Patents
Network public opinion information intelligent processing method and system based on multi-rule association analysis Download PDFInfo
- Publication number
- CN112199583B CN112199583B CN202011003073.8A CN202011003073A CN112199583B CN 112199583 B CN112199583 B CN 112199583B CN 202011003073 A CN202011003073 A CN 202011003073A CN 112199583 B CN112199583 B CN 112199583B
- Authority
- CN
- China
- Prior art keywords
- engine
- label
- data
- engines
- labels
- 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.)
- Active
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
- G06F16/90—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
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2462—Approximate or statistical queries
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2465—Query processing support for facilitating data mining operations in structured databases
Abstract
The invention discloses an online public opinion information intelligent processing method and system based on multi-rule association analysis. The method comprises the following steps: 1) constructing a label system for the selected object, wherein the label system is of a tree structure, a plurality of secondary label nodes are firstly established according to label types, and a plurality of layers of nodes are arranged below each secondary label node; 2) the scheduler sets data types and receiving fields corresponding to the required filtering for each engine according to the configuration of the engines, reads corresponding labels from a label system and distributes the labels to the corresponding engines; 3) the dispatcher sends the data to corresponding engines after receiving the message queue data, and each engine identifies the received data according to the distributed labels and then returns the identification result to the dispatcher; 4) the dispatcher sets corresponding labels for corresponding data according to the identification result, and simultaneously counts the processing information of each engine and stores the processing information into a task counting library; 5) and the task statistic library acquires the current network public opinion according to the statistic information.
Description
Technical Field
The invention relates to an online public opinion information intelligent processing method and system based on multi-rule association analysis, and belongs to the technical field of computer software.
Background
The big data label plays an important application in the present-stage data expression form, and how to combine the traditional label and the present intelligent label together plays an important role in public opinion analysis. The traditional tags are mainly some tags on data attributes, the big data tags are mainly aimed at some characteristic type tags obtained through a data mining technology, and the marking of data objects is completed by establishing the combination based on the traditional tags and intelligent tags and intelligently processing data.
In the intelligent processing part, as data mining and artificial intelligence are developed very fast, the functional points required by public sentiment are different. Therefore, the function system customized based on the user requirement becomes an urgent need for intelligent data processing, and the system integrates various functions and provides the function customized by the user according to the requirement.
Disclosure of Invention
The invention aims to provide a network public opinion information intelligent processing method and system based on multi-rule association analysis, and a multi-level multi-branch label system is constructed aiming at the current and complex mass data of the diversification of public opinion information. When public sentiment analysis is carried out, labeling service is provided, and quick and high-quality data service is provided for the public sentiment analysis. The invention intelligently identifies and marks different types of labels through an intelligent processing engine.
The technical scheme of the invention is as follows:
an online public opinion information intelligent processing method based on multi-rule association analysis comprises the following steps:
1) constructing a label system for the selected object, wherein the label system is of a tree structure, a plurality of secondary label nodes are firstly established according to label types, a plurality of layers of nodes are arranged below each secondary label node, the ith layer of node is a father node of the adjacent (i + 1) th layer of node, and the (i + 1) th layer of node is a child node of the adjacent ith layer of node; each label only belongs to one type of secondary label node, one node has one or more child nodes, and each child node has one or more father nodes; the label types comprise general class labels, intelligent labels and business labels;
2) when the scheduler is started, reading engine configuration in a configuration library, setting data types and receiving fields corresponding to filtering requirements for each engine according to the engine configuration, and reading corresponding labels from a label system and distributing the labels to the corresponding engines;
3) the scheduler sends the data to corresponding engines according to formats defined by the engines after receiving the message queue data, and each engine identifies the received data according to the distributed labels and then returns the identification result to the scheduler; aiming at an engine with a dependency relationship, namely the input data of an engine i is the identification result of the engine j, monitoring the data task state of the engine j, sending the identification result of the engine j to the engine i and a scheduler after the task of the engine j is completed, and then returning the identification result to the scheduler by the engine i;
4) the scheduler sets corresponding labels for corresponding data according to the identification result, and simultaneously counts the processing information of each engine and stores the statistical information into a task statistical library;
5) and the task statistic library acquires the current network public opinion according to the statistic information.
Further, the scheduler sets a corresponding label and upper and lower labels of the label for the corresponding data according to the identification result.
Further, the configuration library is used for indicating the needed engines, the association relationship among the engines, the number of the engines configured when the engines are started, the label definition corresponding to the engines, the data types received by the engines, the receiving fields, the fields returned after processing and the transmission protocol.
Further, the method for the task statistics library to obtain the current internet public sentiment according to the statistical information comprises the following steps: acquiring data associated with the target data according to the label of the target data and the label hierarchy expanded downwards to obtain an associated data set of the target data; and then, identifying different stages of the event corresponding to the target data and public opinion propagation conditions according to the associated data set.
Further, the label types include general class labels, smart labels and business labels.
Further, the selected objects are individuals or groups.
An online public opinion information intelligent processing system based on multi-rule association analysis is characterized by comprising a label system, a configuration library, a task statistic library, a scheduler and a plurality of engines; wherein the content of the first and second substances,
the label system is of a tree structure, wherein a plurality of secondary label nodes are established according to label types, a plurality of layers of nodes are arranged below each secondary label node, the ith layer of nodes are father nodes of adjacent (i + 1) th layer of nodes, and the (i + 1) th layer of nodes are child nodes of the adjacent ith layer of nodes; each label only belongs to one type of secondary label node, one node has one or more child nodes, and each child node has one or more father nodes; the label types comprise general class labels, intelligent labels and business labels;
the scheduler is used for reading the engine configuration in the configuration library during starting, setting the data type and the receiving field corresponding to the required filtering for each engine according to the engine configuration, and reading the corresponding label from the label system and distributing the label to the corresponding engine; after receiving the message queue data, sending the data to a corresponding engine according to a format defined by each engine; setting corresponding labels for corresponding data according to the recognition results of the engines, and meanwhile, counting the processing information of each engine and storing the counting information into a task counting library;
the engine is used for identifying the received data according to the distributed labels and then returning an identification result to the scheduler; aiming at an engine with a dependency relationship, namely the input data of an engine i is the identification result of the engine j, monitoring the data task state of the engine j, sending the identification result of the engine j to the engine i and a scheduler after the task of the engine j is completed, and then returning the identification result to the scheduler by the engine i;
and the task statistic library is used for storing the statistic information and acquiring the current network public opinion according to the statistic information.
Positive effects of the invention
The network public opinion information intelligent processing framework based on the label system can perfectly complete intelligent processing under the condition of network public opinion big data through the label system and the intelligent scheduling system. All engine scheduling configurations are achieved, intelligent marking of result data is achieved, and efficient data support is provided for future public opinion analysis.
Drawings
Fig. 1 is a schematic diagram of a tag type.
Fig. 2 is an exemplary diagram of a label system.
Fig. 3 is an exemplary diagram of a data marking process.
Fig. 4 is an exemplary graph of the result collection.
FIG. 5 is a flow chart of the present invention.
Detailed Description
In order to make the technical solutions in the embodiments of the present invention better understood and make the objects, features, and advantages of the present invention more comprehensible, the technical core of the present invention is described in further detail below with reference to the accompanying drawings.
The tag system is formed by combining a traditional tag and a big data tag. According to the new label system, labels are mainly defined as general class labels, intelligent labels and service labels according to functional dimensions.
The general class label is a pile of special intelligent labels or special feature labels, for example, when a person is used as a general label, the person may have some special attribute labels, for example, the sex, age and height of the person are all labels, and some special feature labels, for example, the emotional tendency and the professional ability evaluation of the person, which are combined to form a portrait of the person. Of course, a summary class label may also contain multiple summary class labels. For example, when a team is a tag, he may have a generalized class tag of multiple people below him. Such as leaders, financial mains, design mains, etc.
The service tags are tags which need to be customized for the system and some tags which are defined for the message information attribute in the big data, and the tags are important pillars for assisting the service function and public opinion analysis. Especially for text content in big data, label filtering of keyword combination is an important information filtering means in public opinion analysis. Such as keyword filtering labels (a | C) & B. There are some other attribute labels, such as region labels corresponding to regions contained in the data, etc., and certainly, according to the service requirement, the service label may also be a combination of some service labels and smart labels, such as information filtering labels including region label (beijing) and keyword label (zhongguancun & beiqianqi).
The intelligent tag is a tag generated for data through data mining and artificial intelligence, and comprises classification, emotion and the like of text data, voiceprint, face, scene recognition and the like of media type data. The label identified by data mining and artificial intelligence in public opinion analysis is an important part for assisting public opinion analysis, and the identification accuracy is also the key for determining the public opinion analysis effect.
The labels are divided into four types according to the attribute dimension of the labels, namely keyword type labels, common type rules, range type rules and combination type rules.
The keyword class labels are also mentioned above, and in network public opinion analysis, information filtering plays an important role in public opinion discovery and analysis; the filtering for keyword relational expressions is one of the main tag types, which is mainly used for some service tags that the user can customize.
The keyword filtering is a Trie tree, also called word search tree or key tree, which is a tree structure and a variety of hash tree. Typical applications are for counting and ordering large numbers of strings (but not limited to strings), and are therefore often used by search engine systems for text word frequency statistics.
The keyword filtering is introduced above, when filtering the keyword expression, the keyword expression is firstly broken into keywords according to the expression, and finally, according to the Trie tree comparison, the matched keyword is 1, the unmatched keyword is 0, and according to the expression, bit operation is performed, and finally, the hit tag of 1 and the miss tag of 0 are calculated.
The common type of tags are tags obtained for attributes or through data mining or artificial intelligence, and the tags are directly obtained by a data processing program, do not contain rule matching of keywords, and do not contain the range of numbers. Such tags are mainly attribute tags for some information and smart tags for data mining and artificial intelligence.
The scope type label is also mainly based on some attribute labels or some service labels capable of calculating the scope on the intelligent label, such as children, teenagers, the elderly and the like which may be packaged on the basis of an age label. East, west, north, south, etc. encapsulated on the geographic label.
The combined type label is a type of service label, and the label of the type is a complex expression label generated by combining several labels together according to service requirements.
The intelligent processing part mainly comprises a configuration library, a label system, a cache library, a task state library, a scheduler, result aggregation and an engine.
The configuration library is used for marking engines required by the system, the association relationship among the engines, the number of the engines configured when the engines are started, the label definition corresponding to the engines, the data types received by the system, the fields required, the fields returned after processing, the transmission protocol and the like.
The label system is used for defining the processing result of the engine, after the data flows through the engine, the engine can generate corresponding result data, and the scheduler finds the corresponding label according to the result and marks the data.
Label System building Process
The label system construction is divided into label type construction and label construction.
The label type is constructed, firstly, the labels are classified into three types, namely general labels, intelligent labels and business labels. These three major categories are in a type system, and a user can define a label type (secondary label type) by himself for the intelligent processing service to mark for use, as shown in fig. 1.
The label system is constructed in a tree structure, and each label only belongs to one secondary label type. When the engine is used, a user can search a certain type of label and check the parent nodes of the upper N layers of the node or the child nodes of the lower N layers of the node. The above architecture can be user defined. An example is shown in figure 2.
The cache library is used for caching data in processing, for each engine, after data processing is completed, results are cached in the cache library, and when the results are collected, the data processed by all the engines are stored in the cache library and then are removed from the cache library.
The task state library sends each batch of data to each engine as a task after the scheduler takes out the data from the queue, so that the processing capacity statistics and the completion state of each engine can be stored in the task state library. For front-end display and result collection.
And the scheduler distributes the data to each engine after receiving the message queue data, waits for the engine to return a result, and marks the data according to a label system defined in advance after the engine returns the result. And stores the results in a cache library. And meanwhile, counting the performance of each engine and storing result data into a task state counting library. The data is labeled as shown in fig. 3, for example.
When the scheduler schedules the face engine, XXX face labels can be obtained according to the type of the secondary labels, when media data flow through the face recognition engine, matched face labels are returned to the scheduler, and the scheduler marks the labels corresponding to XXX on the data matched with XXX. Certainly, when the label is obtained, an upper-level label or a lower-level label may be obtained, and at this time, whether to mark the associated upper-level label or lower-level label may be determined.
The engine is a program for processing data, and processes the data stream after the data stream has been streamed to generate a field generated by the program and return the field to the scheduler.
And when the task is completed, storing the result set processed by the engine into the database, and retransmitting the data to the engine for processing aiming at the engine with failure. A decision configurable in the result collection distributes the data to different data sources by type, by filter condition.
The result collection program reads the completed tasks, reads the processed data in the cache library, clears the completed data, reads the shunting configuration, and distributes the data to different data sources. Data distribution refers to different service demand parties, filtering conditions can be set for data fields (including label fields) to obtain desired data contents, all data can be represented when the filtering conditions are not set, and only the fields can be set to be received and fixed, so that storage resources are used at a minimum.
The overall process flow is shown in FIG. 5.
And when the scheduler is started, reading the configuration library, reading the configuration of the engine, filtering the data type required by the corresponding engine and the field received by the engine according to the configuration of the engine, reading the tag from the tag system according to the configuration of the engine, and after the initialization is finished. And sending the data to the engines according to the format defined by the engines, waiting for the processing results of the engines, mapping the engine results and the defined labels to obtain corresponding data labels for marking the data after the results are returned by the engines, and counting the processing information of each engine and storing the processing information into a task counting library.
When public opinion analysis is carried out, a user can configure labels and label hierarchies expanded downwards to acquire associated data. Therefore, the most similar data set can be obtained during public opinion analysis. The method can judge the different stages of the public sentiment such as the induction period, the incubation period, the occurrence period, the development period, the high tide period, the processing period, the rest period, the feedback period and the like more accurately, and meanwhile, the propagation condition of the public sentiment is thinner, and powerful data support is provided for the accuracy of public sentiment analysis.
For an engine with dependency, the data task state needs to be monitored, and the scheduling of the dependency engine can be continuously completed after the dependency engine is completed.
And a result summarizing and monitoring task statistic library, and storing the cache result into a database when a completed task is available. And when the task fails, resending the failed task according to the engine which fails in processing.
Although specific details of the invention, algorithms and figures are disclosed for illustrative purposes, these are intended to aid in the understanding of the contents of the invention and the implementation in accordance therewith, as will be appreciated by those skilled in the art: various substitutions, changes and modifications are possible without departing from the spirit and scope of the present invention and the appended claims. The invention should not be limited to the preferred embodiments and drawings disclosed herein, but rather should be defined only by the scope of the appended claims.
Claims (7)
1. An online public opinion information intelligent processing method based on multi-rule association analysis comprises the following steps:
1) constructing a label system for the selected object, wherein the label system is of a tree structure, a plurality of secondary label nodes are firstly established according to label types, a plurality of layers of nodes are arranged below each secondary label node, the ith layer of node is a father node of the adjacent (i + 1) th layer of node, and the (i + 1) th layer of node is a child node of the adjacent ith layer of node; each label only belongs to one type of secondary label node, one node has one or more child nodes, and each child node has one or more father nodes; the label types comprise general class labels, intelligent labels and business labels; the selected objects are a plurality of individuals or groups;
2) when the scheduler is started, reading engine configuration in a configuration library, setting data types and receiving fields corresponding to filtering requirements for each engine according to the engine configuration, and reading corresponding labels from a label system and distributing the labels to the corresponding engines;
3) the scheduler sends the data to corresponding engines according to formats defined by the engines after receiving the message queue data, and each engine identifies the received data according to the distributed labels and then returns the identification result to the scheduler; aiming at an engine with a dependency relationship, namely the input data of an engine i is the identification result of the engine j, monitoring the data task state of the engine j, sending the identification result of the engine j to the engine i and a scheduler after the task of the engine j is completed, and then returning the identification result to the scheduler by the engine i;
4) the scheduler sets corresponding labels for corresponding data according to the identification result, and simultaneously counts the processing information of each engine and stores the statistical information into a task statistical library;
5) and the task statistic library acquires the current network public opinion according to the statistic information.
2. The method of claim 1, wherein the scheduler sets a corresponding tag and upper and lower level tags of the tag for corresponding data according to the recognition result.
3. The method of claim 1, wherein the configuration library is used to indicate the required engines, the association relationship between the engines, the number of engines configured when starting the engines, the label definition corresponding to the engines, the type of data received by the engines, the received fields, the fields returned after processing, and the transmission protocol.
4. The method of claim 1, wherein the task statistics base obtains the current internet public opinion according to the statistical information by: acquiring data associated with the target data according to the label of the target data and the label hierarchy expanded downwards to obtain an associated data set of the target data; and then, identifying different stages of the event corresponding to the target data and public opinion propagation conditions according to the associated data set.
5. An online public opinion information intelligent processing system based on multi-rule association analysis is characterized by comprising a label system, a configuration library, a task statistic library, a scheduler and a plurality of engines, wherein the label system, the configuration library, the task statistic library, the scheduler and the engines are constructed for selected objects, and the selected objects are a plurality of individuals or groups; wherein the content of the first and second substances,
the label system is of a tree structure, wherein a plurality of secondary label nodes are established according to label types, a plurality of layers of nodes are arranged below each secondary label node, the ith layer of nodes are father nodes of adjacent (i + 1) th layer of nodes, and the (i + 1) th layer of nodes are child nodes of the adjacent ith layer of nodes; each label only belongs to one type of secondary label node, one node has one or more child nodes, and each child node has one or more father nodes; the label types comprise general class labels, intelligent labels and business labels;
the scheduler is used for reading the engine configuration in the configuration library during starting, setting the data type and the receiving field corresponding to the required filtering for each engine according to the engine configuration, and reading the corresponding label from the label system and distributing the label to the corresponding engine; after receiving the message queue data, sending the data to a corresponding engine according to a format defined by each engine; setting corresponding labels for corresponding data according to the recognition results of the engines, and meanwhile, counting the processing information of each engine and storing the counting information into a task counting library;
the engine is used for identifying the received data according to the distributed labels and then returning an identification result to the scheduler; aiming at an engine with a dependency relationship, namely the input data of an engine i is the identification result of the engine j, monitoring the data task state of the engine j, sending the identification result of the engine j to the engine i and a scheduler after the task of the engine j is completed, and then returning the identification result to the scheduler by the engine i;
and the task statistic library is used for storing the statistic information and acquiring the current network public opinion according to the statistic information.
6. The internet public opinion information intelligent processing system according to claim 5, wherein the scheduler sets a corresponding label and upper and lower labels of the label for corresponding data according to the recognition result.
7. The system of claim 5, wherein the configuration library is used for indicating the required engines, the association relationship between the engines, the number of the engines configured when the engines are started, the label definition corresponding to the engines, the data types received by the engines, the received fields, the fields returned after processing and the transmission protocol.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011003073.8A CN112199583B (en) | 2020-09-22 | 2020-09-22 | Network public opinion information intelligent processing method and system based on multi-rule association analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011003073.8A CN112199583B (en) | 2020-09-22 | 2020-09-22 | Network public opinion information intelligent processing method and system based on multi-rule association analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112199583A CN112199583A (en) | 2021-01-08 |
| CN112199583B true CN112199583B (en) | 2021-11-23 |
Family
ID=74015921
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011003073.8A Active CN112199583B (en) | 2020-09-22 | 2020-09-22 | Network public opinion information intelligent processing method and system based on multi-rule association analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112199583B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108170841A (en) * | 2018-01-16 | 2018-06-15 | 深圳市中易科技有限责任公司 | A kind of mobile phone the analysis of public opinion decision-making technique based on information value |
| CN109992661A (en) * | 2019-03-05 | 2019-07-09 | 广发证券股份有限公司 | A kind of intelligent public sentiment monitoring method and system towards securities industry |
| CN110990565A (en) * | 2019-11-20 | 2020-04-10 | 广州商品清算中心股份有限公司 | Extensible text analysis system and method for public sentiment analysis |
| CN111538888A (en) * | 2020-06-05 | 2020-08-14 | 国网山东省电力公司检修公司 | Network public opinion intensity evolution analysis system based on active monitoring engine and big data |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7644057B2 (en) * | 2001-01-03 | 2010-01-05 | International Business Machines Corporation | System and method for electronic communication management |
-
2020
- 2020-09-22 CN CN202011003073.8A patent/CN112199583B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108170841A (en) * | 2018-01-16 | 2018-06-15 | 深圳市中易科技有限责任公司 | A kind of mobile phone the analysis of public opinion decision-making technique based on information value |
| CN109992661A (en) * | 2019-03-05 | 2019-07-09 | 广发证券股份有限公司 | A kind of intelligent public sentiment monitoring method and system towards securities industry |
| CN110990565A (en) * | 2019-11-20 | 2020-04-10 | 广州商品清算中心股份有限公司 | Extensible text analysis system and method for public sentiment analysis |
| CN111538888A (en) * | 2020-06-05 | 2020-08-14 | 国网山东省电力公司检修公司 | Network public opinion intensity evolution analysis system based on active monitoring engine and big data |
Non-Patent Citations (2)
| Title |
|---|
| 多元与智能:跨媒体网络舆情数据的分析与处理;段峰峰等;《传媒观察》;20200110(第01期);全文 * |
| 政府网络舆情监测系统的功能分析;肖慧华;《科技广场》;20170330(第03期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112199583A (en) | 2021-01-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111782965B (en) | Intention recommendation method, device, equipment and storage medium | |
| WO2022134794A1 (en) | Method and apparatus for processing public opinions about news event, storage medium, and computer device | |
| US9413702B2 (en) | Method and apparatus for distributing published messages | |
| JP2007507775A (en) | How to cluster and query media items | |
| CN112148889A (en) | Recommendation list generation method and device | |
| CN111191125A (en) | Data analysis method based on tagging | |
| JafariAsbagh et al. | Clustering memes in social media streams | |
| US9069880B2 (en) | Prediction and isolation of patterns across datasets | |
| US10853579B2 (en) | Mixed-initiative dialog automation with goal orientation | |
| CN112396462B (en) | Crowd circling method and device based on click house | |
| CN107633022A (en) | Personnel's portrait analysis method, device and storage medium | |
| WO2022188646A1 (en) | Graph data processing method and apparatus, and device, storage medium and program product | |
| CN112925899B (en) | Ordering model establishment method, case clue recommendation method, device and medium | |
| CN112199583B (en) | Network public opinion information intelligent processing method and system based on multi-rule association analysis | |
| CN116955856A (en) | Information display method, device, electronic equipment and storage medium | |
| US9336280B2 (en) | Method for entity-driven alerts based on disambiguated features | |
| CN116883035A (en) | Service matching method based on user grouping statistics | |
| US11314793B2 (en) | Query processing | |
| CN116089723A (en) | Recommendation system recommendation method and device | |
| US20220156228A1 (en) | Data Tagging And Synchronisation System | |
| US11829379B2 (en) | Methods and systems of a matching platform for entitites | |
| CN113672603A (en) | Multi-source heterogeneous electric power big data automatic label implementation method and system | |
| CN109408704B (en) | Fund data association method, system, computer device and storage medium | |
| CN113762703A (en) | Method and device for determining enterprise portrait, computing equipment and storage medium | |
| JP2000259529A (en) | Personal profile managing device and storage medium |
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 |