CN112015987B - Potential customer recommendation system and method based on enterprise tags - Google Patents
Potential customer recommendation system and method based on enterprise tags Download PDFInfo
- Publication number
- CN112015987B CN112015987B CN202010883727.4A CN202010883727A CN112015987B CN 112015987 B CN112015987 B CN 112015987B CN 202010883727 A CN202010883727 A CN 202010883727A CN 112015987 B CN112015987 B CN 112015987B
- Authority
- CN
- China
- Prior art keywords
- client
- data
- user
- enterprise
- recommendation
- 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
- G06F16/10—File systems; File servers
- G06F16/18—File system types
- G06F16/182—Distributed file systems
-
- 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
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- 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
- G06F16/951—Indexing; Web crawling techniques
-
- 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
- G06F16/958—Organisation or management of web site content, e.g. publishing, maintaining pages or automatic linking
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/216—Parsing using statistical methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/237—Lexical tools
- G06F40/247—Thesauruses; Synonyms
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/284—Lexical analysis, e.g. tokenisation or collocates
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/10—Office automation; Time management
- G06Q10/103—Workflow collaboration or project management
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Artificial Intelligence (AREA)
- Business, Economics & Management (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Health & Medical Sciences (AREA)
- Strategic Management (AREA)
- Quality & Reliability (AREA)
- Entrepreneurship & Innovation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Economics (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Probability & Statistics with Applications (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a potential customer recommendation system and a potential customer recommendation method based on enterprise tags, wherein the system comprises the following steps: the system comprises a client information management module, a client label management module, a client recommendation module and a client management module; the recommendation method comprises the steps of data crawling and storing, data cleaning, label extracting and cleaning, label printing by a user, graph data structure construction, calculation of the correlation degree of a client vertex relative to the user vertex and recommendation accuracy, and selection of a trained data model with high accuracy to be deployed on a server for client recommendation. The system and the method disclosed by the invention utilize a distributed big data processing technology to extract the enterprise tags more comprehensively, so that the enterprise is described more comprehensively by using the tags, the relation between the interest tags of the user and the enterprise tags is established more accurately, and the recommendation result is more accurate.
Description
Technical Field
The invention relates to the technical field of customer recommendation information, in particular to a potential customer recommendation system and a potential customer recommendation method based on enterprise tags.
Background
With the development of information technology and the internet, people gradually move from an information-deficient era to an information-overloaded era. In this age, people produce more and more data. But how to make these data work is a great challenge to mine information that users feel interesting.
In recent years, socio-economic speed is slowed down, and enterprises face greater and greater survival pressure, wherein one of the great problems faced by the enterprises is the problem of marketing and customer acquisition. Mining potential customers of an enterprise is an urgent need of current enterprises.
From the technical level, the enterprises are expected to be helped to develop the value of the data; from the development aspect, the enterprise is hoped to be helped to reduce the customer acquisition cost, and the power enterprise breaks through the development bottleneck.
Disclosure of Invention
In order to solve the technical problem, the invention provides a potential customer recommendation system and a potential customer recommendation method based on enterprise tags, so as to achieve the purpose of more accurate recommendation results.
In order to achieve the purpose, the technical scheme of the invention is as follows:
an enterprise tag-based potential customer recommendation system comprising:
a client information management module: the system is used for managing client information, including basic information, legal action, business condition, business risk, enterprise development, intellectual property and news bulletin of the client;
the client label management module: calculating a common label of a client by using a TF-IDF method, and pushing the common label to a user, wherein the user corrects and supplements the label according to own experience;
a client recommendation module: calculating potential customers of the enterprise by using a graph-based recommendation algorithm, and pushing the potential customers to the user;
a client management module: and the users market and track the clients, and record the marketing and tracking results in the client management, verify the accuracy of recommendation through the marketing and tracking results, and reversely optimize the tag system and the recommendation algorithm.
A potential customer recommendation method based on enterprise labels comprises the following steps:
(1) Crawling and storing data: the method comprises the steps that a distributed acquisition technology is adopted to regularly acquire website information of enterprises and store data;
(2) Data cleaning: unifying data formats of different data sources;
(3) Extracting and cleaning labels: extracting labels from the cleaned data by using a text mining technology, and removing meaningless labels;
(4) Labeling by a user: designating one collected enterprise as a user u and the rest as a client k, and enabling the user u to label the client k, wherein the label is selected from the cleaned labels obtained in the step (3), and the user label behavior that the user u labels a to the client k is represented by (u, k, a);
(5) And (3) constructing a graph data structure: randomly dividing a data set into a training set and a testing set, representing label behaviors of users in the training set on a bipartite graph, and defining three different vertexes: adding three edges to a user vertex v (u) corresponding to a user u, a client vertex v (k) corresponding to a client k and a label vertex v (a) corresponding to a label a, adding one edge between the user vertex v (u) and the client vertex v (k), adding one edge between the user vertex v (u) and the label vertex v (a), adding one edge between the client vertex v (k) and the label v (a), and adding 1 to the weight of the edge if the two vertices are connected by the edge;
(6) Calculating the relevance of the client vertex relative to the user vertex:
and calculating the correlation of all the client vertexes relative to the user vertexes by adopting a modified PersonalRank algorithm, wherein the iterative formula of the PersonalRank is as follows:
wherein PR (u) is the correlation of all client vertices relative to u, d is the probability of continuing random walk, in (u) is the set of all clients pointing to u, PR (k) is the correlation of all client vertices relative to k, and out (k) is the set of client vertices in the graph connected with k;
definition of r i The following were used:
r i is the probability of firing of vertex u;
(7) Calculation of recommendation accuracy:
selecting customers with high relevance to the user u as a set R (u), wherein T (u) is a set of customers which are actually labeled by the user u in the test set, and evaluating the accuracy of the personalized recommendation algorithm by using Precision and Recall:
(8) And selecting the trained data model with high accuracy to be deployed on a server for customer recommendation.
In the scheme, the data information of the enterprises collected and stored in the step (1) comprises basic information of the enterprises, legal action, business conditions, business risks, enterprise development, intellectual property rights and news bulletins.
In the scheme, in the step (1), the data of the enterprise is divided into structural data and non-structural data, the structural data is stored by adopting an hbase distributed file system, and the non-structural data is stored in an hdfs distributed system.
In the above scheme, in the step (2), the data cleaning method includes the following steps:
(1) Removing the html label;
(2) Removing the special characters;
(3) Detecting and merging repeated records;
(4) Unifying data of different formats of different data sources into one.
In the scheme, in the step (3), the structural data is directly stored in the hbase, or is stored in the hbase after batch calculation in spark environment; for unstructured data, the tags are extracted using text mining techniques.
In the above scheme, in step (3), the label cleaning method includes:
(1) Removing stop words with high word frequency;
(2) Removing synonyms caused by different root words;
(3) Synonyms due to separators are removed.
In the above scheme, in step (5), the data set is randomly divided into a training set and a testing set according to 9.
Through the technical scheme, the potential customer recommendation system and the recommendation method based on the enterprise tags acquire and clean multi-dimensional data related to enterprises in real time, tag the enterprise data, analyze the data of the clients and the clients interested in the transaction of the users, establish the relation between the interests of the users and the clients through the tags, and recommend the enterprises which are possibly interested by the users to the users so as to reduce the customer acquisition cost of the users. The invention utilizes the distributed big data processing technology to extract the enterprise tags more comprehensively, thereby describing the enterprises more comprehensively by the tags and further establishing the relation between the interest tags of the users and the enterprise tags more accurately.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below.
Fig. 1 is a schematic flow chart of a potential customer recommendation method based on enterprise tags according to an embodiment of the present invention.
Detailed Description
The technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.
The invention provides a potential customer recommendation system based on enterprise labels, which comprises:
the client information management module: the system is used for managing client information, including basic information, legal action, business condition, business risk, enterprise development, intellectual property and news bulletin of the client;
the client label management module: calculating a common label of a client by using a TF-IDF method, pushing the common label to a user, and correcting and supplementing the label by the user according to own experience;
a client recommendation module: calculating potential customers of the enterprise by using a graph-based recommendation algorithm, and pushing the potential customers to the user;
a client management module: and the users carry out marketing tracking on the clients, marketing tracking results are recorded in client management, the accuracy of recommendation is verified through the marketing tracking results, and the tag system and the recommendation algorithm are reversely optimized.
A potential customer recommendation method based on enterprise tags, as shown in fig. 1, includes the following steps:
1. crawling and storing data:
regularly acquiring website information of enterprises every day by adopting a distributed acquisition technology, and storing data;
the collected and stored data information of the enterprise comprises basic information, legal action, business condition, business risk, enterprise development, intellectual property and news bulletin of the enterprise.
The data of the enterprise are divided into structural data and non-structural data, the structural data are stored by adopting an hbase distributed file system, and the non-structural data are stored in an hdfs distributed system.
2. Data cleaning:
data collected from a website is very disordered, a large number of webpage labels and irrelevant data are collected and need to be cleaned, and data formats of different data sources are greatly different and need to be unified. The cleaning and sorting module utilizing the data cleans regularly every day, and the method for cleaning the data comprises the following steps:
(1) Removing the html label;
(2) Removing special characters;
(3) Detecting and merging repeated records;
(4) Unifying data of different formats of different data sources into one.
3. Extracting and cleaning labels:
with the help of business experts, the label of each data dimension is determined. For structural data, some fields can be used as tags and stored in hbase, and some fields need to be calculated in bulk in spark environment and then stored in hbase.
For non-structural line data, extracting tags by using a text mining technology, wherein the used text mining technology comprises the following steps:
data word segmentation, namely segmenting text type data into words by using a third-party word segmentation tool jieba, and then selecting medium labels from the words;
and (4) finding new words, and calculating the possibility of combining characters left and right into new words by using the information entropy.
The label cleaning method comprises the following steps:
(1) Removing stop words with high word frequency;
(2) Removing synonyms caused by different root words;
(3) Synonyms due to separators are removed.
4. Labeling by a user:
and (c) designating one collected enterprise as a user u, and the rest enterprises as customers k, and enabling the user u to label the customers k, wherein the labels are selected from the cleaned labels obtained in the step three, and the user label behaviors that the user u labels a to the customers k are represented by (u, k, a).
5. Construct graph data structure:
the data set is randomly divided into a training set and a testing set according to the following 9. That is, the user's multiple labels for the customer are grouped into either a training set or a test set.
Representing the label behaviors of users in the training set on a bipartite graph, and defining three different vertexes: the method comprises the steps that a user vertex v (u) corresponding to a user u, a client vertex v (k) corresponding to a client k and a label vertex v (a) corresponding to a label a are added, three edges are added in a graph, one edge is added between the user vertex v (u) and the client vertex v (k), one edge is added between the user vertex v (u) and the label vertex v (a), one edge is added between the client vertex v (k) and the label v (a), and if edges are connected between the two vertices, the weight of the edges is increased by 1.
6. Calculating the relevance of the client vertex relative to the user vertex:
and calculating the correlation of all client vertexes relative to the user vertexes by adopting a modified Personalrank algorithm, wherein the iterative formula of the Personalrank is as follows:
PR (u) is the correlation of all client vertices relative to u, d is the probability of continuing random walk, in (u) is the set of all clients pointing to u, PR (k) is the correlation of all client vertices relative to k, and out (k) is the set of client vertices in the graph connected with k;
definition of r i The following were used:
r i is the probability of the start of vertex u.
Using the above iterative formula, the correlation of all the customer vertices with respect to the user vertices can be calculated.
7. Calculation of recommendation accuracy:
selecting a client with high user vertex correlation corresponding to a user u as a set R (u), wherein T (u) is a set of clients actually labeled by the user u in a test set, and evaluating the accuracy of the personalized recommendation algorithm by using Precision and Recall:
8. and selecting a trained data model with high accuracy to be deployed on the server, and providing a model interface for customer recommendation.
The recommendation based on the label has the greatest advantage that the label can be used for recommendation explanation, and the user is difficult to intuitively feel that the recommendation result is reasonable.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (7)
1. A potential customer recommendation method based on enterprise labels adopts a potential customer recommendation system based on enterprise labels, and is characterized in that the recommendation system comprises:
a client information management module: the system is used for managing client information, including basic information, legal action, business condition, business risk, enterprise development, intellectual property and news bulletin of the client;
a client tag management module: calculating a common label of a client by using a TF-IDF method, and pushing the common label to a user, wherein the user corrects and supplements the label according to own experience;
a client recommendation module: calculating potential customers of the enterprise by using a graph-based recommendation algorithm, and pushing the potential customers to the user;
a client management module: the users track the clients in a marketing manner, marketing tracking results are recorded in client management, the accuracy of recommendation is verified through the marketing tracking results, and a tag system and a recommendation algorithm are reversely optimized;
the recommendation method comprises the following steps:
(1) Crawling and storing data: the method comprises the steps that a distributed acquisition technology is adopted to regularly acquire website information of enterprises and store data;
(2) Data cleaning: unifying data formats of different data sources;
(3) Extracting and cleaning labels: extracting tags from the cleaned data by using a text mining technology, and removing meaningless tags;
(4) Labeling by a user: designating one collected enterprise as a user u and the rest as a client k, and enabling the user u to label the client k, wherein the labels are selected from the cleaned labels obtained in the step (3), and the user label behavior that the user u labels a to the client k is represented by (u, k, a);
(5) Construct graph data structure: randomly dividing a data set into a training set and a testing set, representing the label behaviors of users in the training set on a bipartite graph, and defining three different vertexes: adding three edges to a user vertex v (u) corresponding to a user u, a client vertex v (k) corresponding to a client k and a label vertex v (a) corresponding to a label a, adding one edge between the user vertex v (u) and the client vertex v (k), adding one edge between the user vertex v (u) and the label vertex v (a), adding one edge between the client vertex v (k) and the label v (a), and adding 1 to the weight of the edge if the two vertices are connected by the edge;
(6) Calculating the relevance of the client vertex relative to the user vertex:
and calculating the correlation of all client vertexes relative to the user vertexes by adopting a modified Personalrank algorithm, wherein the iterative formula of the Personalrank is as follows:
wherein PR (u) is the correlation of all client vertices relative to u, d is the probability of continuing random walk, in (u) is the set of all clients pointing to u, PR (k) is the correlation of all client vertices relative to k, and out (k) is the set of client vertices in the graph connected with k;
definition of r i The following were used:
r i being the vertex uStarting probability;
(7) Calculation of recommendation accuracy:
selecting customers with high relevance to the user u as a set R (u), wherein T (u) is a set of customers which are actually labeled by the user u in the test set, and evaluating the accuracy of the personalized recommendation algorithm by using Precision and Recall:
(8) And selecting the trained data model with high accuracy to be deployed on the server for customer recommendation.
2. The method of claim 1, wherein the data information of the enterprise collected and stored in the step (1) includes basic information of the enterprise, legal action, business situation, business risk, enterprise development, intellectual property and news bulletin.
3. The method for recommending potential clients based on enterprise tags as claimed in claim 1, wherein in step (1), the data of the enterprise is divided into structural data and non-structural data, the structural data is stored by using hbase distributed file system, and the non-structural data is stored in hdfs distributed system.
4. The enterprise tag-based potential customer recommendation method according to claim 1, wherein in the step (2), the data cleansing method comprises the following steps:
(1) Removing the html label;
(2) Removing special characters;
(3) Detecting and merging repeated records;
(4) Unifying data of different formats of different data sources into one.
5. The enterprise tag-based potential customer recommendation method according to claim 1, wherein in the step (3), the structural data is directly stored in the hbase, or is stored in the hbase after batch computation in spark environment; for unstructured data, the tags are extracted using text mining techniques.
6. The enterprise tag-based potential customer recommendation method according to claim 1, wherein in the step (3), the tag cleaning method comprises:
(1) Removing stop words with high word frequency;
(2) Removing synonyms caused by different root words;
(3) Synonyms due to separators are removed.
7. The method for recommending potential customers based on enterprise labels as claimed in claim 1, wherein in step (5), the data set is randomly divided into a training set and a testing set according to 9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010883727.4A CN112015987B (en) | 2020-08-28 | 2020-08-28 | Potential customer recommendation system and method based on enterprise tags |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010883727.4A CN112015987B (en) | 2020-08-28 | 2020-08-28 | Potential customer recommendation system and method based on enterprise tags |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112015987A CN112015987A (en) | 2020-12-01 |
| CN112015987B true CN112015987B (en) | 2022-10-04 |
Family
ID=73502811
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010883727.4A Active CN112015987B (en) | 2020-08-28 | 2020-08-28 | Potential customer recommendation system and method based on enterprise tags |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112015987B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112861004B (en) * | 2021-02-20 | 2024-02-06 | 中国联合网络通信集团有限公司 | Method and device for determining rich media |
| CN113076478B (en) * | 2021-04-14 | 2022-06-07 | 同济大学 | Technical resource and service recommendation system based on hybrid recommendation algorithm |
| CN113590968A (en) * | 2021-08-10 | 2021-11-02 | 平安普惠企业管理有限公司 | Resource recommendation method and device, computer equipment and storage medium |
| CN116561432B (en) * | 2023-06-27 | 2024-05-03 | 广州钛动科技股份有限公司 | Intelligent employee content data recommendation system |
| CN117312678B (en) * | 2023-10-24 | 2024-04-05 | 亿企查科技有限公司 | Intelligent recommendation method, equipment and storage medium for potential clients based on big data |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015021937A1 (en) * | 2013-08-14 | 2015-02-19 | 腾讯科技（深圳）有限公司 | Method and device for user recommendation |
| CN106383894A (en) * | 2016-09-23 | 2017-02-08 | 深圳市由心网络科技有限公司 | Enterprise supply-demand information matching method and apparatus |
| CN108399191A (en) * | 2018-01-25 | 2018-08-14 | 温州大学 | A kind of personalized recommendation method of information on bidding |
| CN110059246A (en) * | 2019-03-15 | 2019-07-26 | 安徽省优质采科技发展有限责任公司 | Intelligent match system |
| CN110647683A (en) * | 2019-09-17 | 2020-01-03 | 北京邮电大学 | Information recommendation method and device |
| CN111079004A (en) * | 2019-12-06 | 2020-04-28 | 成都理工大学 | Three-part graph random walk recommendation method based on word2vec label similarity |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| SG10201500683TA (en) * | 2015-01-28 | 2016-08-30 | Crayon Data Pte Ltd | Method and system for automatically recommending business prospects |
-
2020
- 2020-08-28 CN CN202010883727.4A patent/CN112015987B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015021937A1 (en) * | 2013-08-14 | 2015-02-19 | 腾讯科技（深圳）有限公司 | Method and device for user recommendation |
| CN106383894A (en) * | 2016-09-23 | 2017-02-08 | 深圳市由心网络科技有限公司 | Enterprise supply-demand information matching method and apparatus |
| CN108399191A (en) * | 2018-01-25 | 2018-08-14 | 温州大学 | A kind of personalized recommendation method of information on bidding |
| CN110059246A (en) * | 2019-03-15 | 2019-07-26 | 安徽省优质采科技发展有限责任公司 | Intelligent match system |
| CN110647683A (en) * | 2019-09-17 | 2020-01-03 | 北京邮电大学 | Information recommendation method and device |
| CN111079004A (en) * | 2019-12-06 | 2020-04-28 | 成都理工大学 | Three-part graph random walk recommendation method based on word2vec label similarity |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112015987A (en) | 2020-12-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112015987B (en) | Potential customer recommendation system and method based on enterprise tags | |
| Olmezogullari et al. | Representation of click-stream datasequences for learning user navigational behavior by using embeddings | |
| US20150317303A1 (en) | Topic mining using natural language processing techniques | |
| US8386487B1 (en) | Clustering internet messages | |
| Olmezogullari et al. | Pattern2Vec: Representation of clickstream data sequences for learning user navigational behavior | |
| US20200074242A1 (en) | System and method for monitoring online retail platform using artificial intelligence | |
| WO2015066805A1 (en) | Systems and methods for behavioral segmentation of users in a social data network | |
| CN107291755B (en) | Terminal pushing method and device | |
| Agarwal et al. | Approximate incremental big-data harmonization | |
| CN110543474A (en) | User behavior analysis method and device based on full-buried point and potential factor model | |
| CN113297840B (en) | Malicious traffic account detection method, device, equipment and storage medium | |
| CN104915359A (en) | Theme label recommending method and device | |
| CN113886708A (en) | Product recommendation method, device, equipment and storage medium based on user information | |
| CN110795613A (en) | Commodity searching method, device and system and electronic equipment | |
| Medvet et al. | Brand-related events detection, classification and summarization on twitter | |
| CN118210875A (en) | Knowledge retrieval method and knowledge base management platform based on large language model | |
| US20160371725A1 (en) | Campaign optimization system | |
| US20150235138A1 (en) | System, method, and storage medium for generating hypotheses in data sets | |
| CN113077312A (en) | Hotel recommendation method, system, equipment and storage medium | |
| CN111127057B (en) | Multi-dimensional user portrait recovery method | |
| CN115248855A (en) | Text processing method and device, electronic equipment and computer readable storage medium | |
| CN112948681A (en) | Time series data recommendation method fusing multi-dimensional features | |
| CN104573098B (en) | Extensive object identifying method based on Spark systems | |
| Lee et al. | Event evolution tracking from streaming social posts | |
| CN114490667A (en) | Multidimensional data analysis method and device, electronic equipment and 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 |