CN106599022B - User portrait forming method based on user access data - Google Patents

User portrait forming method based on user access data Download PDF

Info

Publication number
CN106599022B
CN106599022B CN201610935388.3A CN201610935388A CN106599022B CN 106599022 B CN106599022 B CN 106599022B CN 201610935388 A CN201610935388 A CN 201610935388A CN 106599022 B CN106599022 B CN 106599022B
Authority
CN
China
Prior art keywords
user
label
webpage
classifier
access data
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
Application number
CN201610935388.3A
Other languages
Chinese (zh)
Other versions
CN106599022A (en
Inventor
聂琳
林倞
王青
罗思伟
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
National Sun Yat Sen University
Original Assignee
National Sun Yat Sen University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by National Sun Yat Sen University filed Critical National Sun Yat Sen University
Priority to CN201610935388.3A priority Critical patent/CN106599022B/en
Publication of CN106599022A publication Critical patent/CN106599022A/en
Application granted granted Critical
Publication of CN106599022B publication Critical patent/CN106599022B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

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
    • G06F16/35—Clustering; Classification
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—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
    • 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

Abstract

The method provided by the invention utilizes a crawler tool, an extraction algorithm and a Chinese word segmentation method to acquire and automatically process the contents in the webpage, has higher intelligent and automatic degrees, and well solves the defects of the prior art. The method provided by the invention utilizes a machine learning method to learn the characteristics of the user so as to show the behavior preference of the user such as life, shopping and the like.

Description

user portrait forming method based on user access data
Technical Field
the invention relates to the technical field of computers, in particular to a user portrait forming method based on user access data.
Background
big data generally refers to a data set which cannot be captured, managed and processed by a conventional software tool within a certain time range, and is information assets which can be adapted to mass, high growth rate and diversification by having stronger decision making power, insight discovery power and flow optimization capability only through a new processing mode.
The user portrait is an important technical application generated in the big data era, and the aim of the method is to establish multidimensional descriptive label attributes aiming at users, so that real personal characteristics of the users in various aspects are outlined by using the label attributes, further, the user demand can be explored by using the user portrait, the user preference is analyzed, and more efficient and more targeted information transmission and user experience closer to personal habits are provided for the users by matching the user portrait.
At present, the formation of user portraits is generally applied to electronic commerce websites, news recommendation systems and the like, and aims to recommend interested commodities or news to users more accurately and improve user experience.
The data for creating the user representation generally comprises data in real life of the user and network behavior data. The data in real life comprises basic information of the user such as name, sex, age, sports preference and the like, and the network behavior data comprises behavior records of webpage access, games, music listening, movie watching, social contact and the like of the user in the Internet world.
Collaborative Filtering (english: Collaborative Filtering), as a classic method of a user portrayal and recommendation system, is to simply recommend information of interest to a user by using the preferences of groups with mutual interests and experiences, and individuals give a considerable degree of response (such as scoring) to the information through a Collaborative mechanism and record the response to filter the information, so as to help others to filter the information.
Content-based recommendations are recommendations based on the attributes (feature vectors) of the content itself. Feature extraction (vectorization) is required to be carried out on a product, a preference document of a user is established according to historical information of the user, and the preference document can be regarded as a user portrait. Based on the representation, the commodity and service suitable for the user can be found for recommendation.
However, the above methods all need to label the webpage data manually in specific implementation, and the processing efficiency is low.
Disclosure of Invention
The invention provides a user portrait forming method based on user access data for solving the problems of the prior art, the method does not need to label webpage data in the concrete implementation process, has high processing efficiency, and can learn the characteristics of a user by utilizing a machine learning method so as to show behavior preference of the user such as life, shopping and the like.
In order to realize the purpose, the technical scheme is as follows:
A method of user representation creation based on user access data, comprising the steps of:
S1, filtering access data of a user, and filtering irrelevant request links in the access data to obtain relevant access links;
S2, capturing a webpage corresponding to the relevant access link by using a crawler tool, and then extracting text information in the captured webpage by using an extraction algorithm;
S3, performing word segmentation processing on the extracted text information by using a Chinese word segmentation method, wherein a word list obtained after the word segmentation processing is performed on the text information of each webpage is stored in a document;
S4, performing Word segmentation processing on a corpus disclosed on a network, and then training Word vectors Word2Vec by using a Word vector technology based on the corpus subjected to Word segmentation processing to obtain distributed expression of Chinese words;
S5, creating a Doc2Vec model, initializing the Doc2Vec model by using Word vector Word2Vec, then respectively inputting a vocabulary list in each document into the Doc2Vec model, training the Doc2Vec model by using the vocabulary list in the document, and outputting the Doc2Vec model as distributed expression of a webpage corresponding to the document;
s6, for each label, training a judgment classifier for judging whether the distributed expression carries the label or not;
S7, respectively inputting the distributed expression of each webpage in the step S5 into a judgment classifier of each label, and if the output of the judgment classifier of the label is positive, indicating that the webpage of the user accesses the attribute with the label; and if the output of the classifier is judged to be negative, the attribute that the user does not have the label is indicated in the webpage access.
Preferably, in step S1, the CSS request link, the picture resource request link, and the js script resource request link in the access data are filtered out.
Preferably, the irrelevant access is filtered through a regular expression in the step S1.
Preferably, in the step S2, the text information in the crawled webpage is extracted by using an extraction algorithm based on text density.
Preferably, in step S2, a hash table is created to store the crawled web pages.
Preferably, in step S6, the judgment classifier includes a trained convolutional neural network and a logistic regression binary classifier, an output end of the convolutional neural network is connected to an input end of the logistic regression binary classifier, the convolutional neural network is used to classify the label, and the logistic regression binary classifier is used to output a classification result of the convolutional neural network.
Compared with the prior art, the invention has the beneficial effects that:
The method provided by the invention utilizes a crawler tool, an extraction algorithm and a Chinese word segmentation method to acquire and automatically process the contents in the webpage, has higher intelligent and automatic degrees, and well solves the defects of the prior art. The method provided by the invention utilizes a machine learning method to learn the characteristics of the user so as to show the behavior preference of the user such as life, shopping and the like.
Drawings
FIG. 1 is a diagram illustrating a first test result.
FIG. 2 is a diagram illustrating a second test result.
Fig. 3(a) and (b) are a schematic diagram of a test result three and a schematic diagram of a test result four.
Fig. 4 is a diagram showing a test result five.
Fig. 5 is a diagram showing a sixth test result.
FIG. 6 is a diagram of training a decision classifier.
Detailed Description
The drawings are for illustrative purposes only and are not to be construed as limiting the patent;
The invention is further illustrated below with reference to the figures and examples.
example 1
The method provided by the invention mainly comprises the following steps:
and in the first step, unnecessary links are filtered and are gathered and sorted by taking a user as a unit. In the step, unnecessary links such as the open source js library resource access request and the external picture resource access request are filtered out, and the link which the user actually wants to access is obtained. The links and access times for each user are then stored in chronological order.
And secondly, compiling a web crawler tool and capturing all linked web title and content of the user. Because a user has many pages to access, in order to avoid repeatedly fetching a certain page, a hash table needs to be established to store the link of the fetched page, so that repeated fetching can be avoided.
In the crawling process, in order to avoid the anti-crawler mechanism of some websites, several strategies need to be utilized, which are respectively: disguise the user agent, use the agent and avoid continuous access to the same website, etc.
And thirdly, extracting the text content of each webpage from the webpage content by using an open-source webpage text extraction algorithm such as cx-extrator. At present, the web page content generally contains a certain amount of contents which are irrelevant to the page theme, such as advertisements, website directory navigation and the like.
And fourthly, converting each document (title and text) into a list consisting of a series of vocabularies by using a Chinese word segmentation technology for the text and the title of each webpage.
And step five, collecting a corpus on a network, such as the corpus provided by Chinese Wikipedia, segmenting words by the method of step 4, and then training Word vectors Word2Vec of the words by using a Word vector technology. Here, the distributed expression of the trained vocabulary is a Word vector Word2Vec which can express the part of speech, the meaning and the relevance with other words to a certain extent by adopting unsupervised learning and using the skip-gram technology.
As shown in FIG. 1 and FIG. 2, two words with the most similar part of speech and meaning are tested, namely "ok" and "beauty" respectively. Given "may" the most similar words are "can", given "beauty" the similar words are "handsome", "sister".
As further shown in fig. 3(a), 3(b), and 4, when the similarity between a subject vocabulary and a series of related or unrelated words is tested, it is obvious that the score of words related to the subject vocabulary is significantly higher than the score of words not related.
finally, as shown in FIG. 5, the relationship between the vocabulary vectors:
The highest scoring answer from naobao-china-? -usa is ebay.
From the above examples, it can be seen that the power of word2vec learns not only vocabulary similarity but also relationships between vocabularies. So a trained word2vec can be used.
And sixthly, because the access time of each link is stored in the step (1), the access sequence of the links can be obtained according to a certain rule according to the information, and the webpage access documents of each user are sorted according to the time sequence.
and seventhly, initializing a Doc2Vec model by using the Word vector Word2Vec obtained in the fifth step, training a Word list obtained from each document by using a skip gram to obtain distributed expression of the text, wherein the document vector represents information of a document formed by a series of words, generally speaking, a certain theme, a certain emotion and a certain class of commodities.
In this step, a convolutional neural network CNN is used, the assumption is made that a user U corresponds to N different Doc2 Vecs, which are respectively represented as Doc 1 and Doc 2 … Doc N, and a user label vector is a 0-1 vector with the length of N tag, and for each label, a convolutional neural network is trained for classifying the label.
for example: doc2Vec is a 200 dimensional vector, then the input to the net is N x 200. One convolution kernel is set to i x 200, i representing that the convolution kernel convolves i vectors at a time. If n convolution kernels are set at the input level, the total convolution kernel is one kernel of n × i × 200; thus, the output resulting from the convolution of the input layers is a matrix of N x (N-i + 1). Then k-max poling is used to get vectors of fixed length k x n. A fully connected layer is added after this vector, followed by 1 logistic regression binary classifier. As shown in fig. 6.
Thus, a judgment classifier of the label is obtained. Assuming a total of m labels, only m classifiers need to be trained in the same way.
And step nine, in the use stage, for a single user, obtaining all document vectors of the user, and then obtaining m binary classifiers obtained in step 8, wherein the obtained label with a positive output is the label of the user, and according to the score of the last classifier, the higher the score is, the more obvious the label is at the user.
it should be understood that the above-described embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not intended to limit the embodiments of the present invention. Other variations and modifications will be apparent to persons skilled in the art in light of the above description. And are neither required nor exhaustive of all embodiments. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention should be included in the protection scope of the claims of the present invention.

Claims (6)

1. A user portrait forming method based on user access data is characterized in that: the method comprises the following steps:
s1, filtering access data of a user, and filtering irrelevant request links in the access data to obtain relevant access links;
S2, capturing a webpage corresponding to the relevant access link by using a crawler tool, and then extracting text information in the captured webpage by using an extraction algorithm;
S3, performing word segmentation processing on the extracted text information by using a Chinese word segmentation method, wherein a word list obtained after the word segmentation processing is performed on the text information of each webpage is stored in a document;
S4, performing Word segmentation processing on a corpus disclosed on a network, and then training Word vectors Word2Vec by using a Word vector technology based on the corpus subjected to Word segmentation processing to obtain distributed expression of Chinese words;
s5, creating a Doc2Vec model, initializing the Doc2Vec model by using Word vector Word2Vec, then respectively inputting a vocabulary list in each document into the Doc2Vec model, training the Doc2Vec model by using the vocabulary list in the document, and outputting the Doc2Vec model as distributed expression of a webpage corresponding to the document;
S6, for each label, training a judgment classifier for judging whether the distributed expression carries the label or not;
s7, respectively inputting the distributed expression of each webpage in the step S5 into a judgment classifier of each label, and if the output of the judgment classifier of the label is positive, indicating that the webpage of the user accesses the attribute with the label; and if the output of the classifier is judged to be negative, the attribute that the user does not have the label is indicated in the webpage access.
2. A user representation creation method in accordance with claim 1, wherein: in step S1, the CSS request link, the picture resource request link, and the js script resource request link in the access data are filtered out.
3. A user representation creation method in accordance with claim 2 wherein: irrelevant accesses are filtered through regular expressions in the step S1.
4. a user representation creation method in accordance with claim 1, wherein: in step S2, text information in the crawled web page is extracted using an extraction algorithm based on text density.
5. A user representation creation method in accordance with claim 1, wherein: in step S2, a hash table is created to store the links of the pages that have been crawled.
6. A user representation creation method in accordance with claim 1, wherein: in the step S6, the judgment classifier includes a trained convolutional neural network and a logistic regression binary classifier, an output end of the convolutional neural network is connected to an input end of the logistic regression binary classifier, the convolutional neural network is used to classify the label, and the logistic regression binary classifier is used to output a classification result of the convolutional neural network.
CN201610935388.3A 2016-11-01 2016-11-01 User portrait forming method based on user access data Active CN106599022B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610935388.3A CN106599022B (en) 2016-11-01 2016-11-01 User portrait forming method based on user access data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610935388.3A CN106599022B (en) 2016-11-01 2016-11-01 User portrait forming method based on user access data

Publications (2)

Publication Number Publication Date
CN106599022A CN106599022A (en) 2017-04-26
CN106599022B true CN106599022B (en) 2019-12-10

Family

ID=58589465

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610935388.3A Active CN106599022B (en) 2016-11-01 2016-11-01 User portrait forming method based on user access data

Country Status (1)

Country Link
CN (1) CN106599022B (en)

Families Citing this family (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107633036A (en) * 2017-09-08 2018-01-26 广州汪汪信息技术有限公司 A kind of microblog users portrait method, electronic equipment, storage medium, system
CN107895024B (en) * 2017-09-13 2021-10-08 同济大学 User model construction method and recommendation method for webpage news classification recommendation
CN110020113B (en) * 2017-09-28 2021-04-20 南京无界家居科技有限公司 Home product prediction method and device based on feature matching
CN107818334A (en) * 2017-09-29 2018-03-20 北京邮电大学 A kind of mobile Internet user access pattern characterizes and clustering method
CN109002459A (en) * 2018-05-30 2018-12-14 珠海市君天电子科技有限公司 A kind of type of merchandise recognition methods of user preferences and device
CN108874941B (en) * 2018-06-04 2021-09-21 成都知道创宇信息技术有限公司 Big data URL duplication removing method based on convolution characteristics and multiple Hash mapping
CN108769440A (en) * 2018-06-06 2018-11-06 北京京东尚科信息技术有限公司 Preposition shunt method and device
CN110717116B (en) * 2018-06-27 2023-12-05 北京京东尚科信息技术有限公司 Link prediction method and system of relational network, equipment and storage medium
CN108521435B (en) * 2018-07-06 2021-09-03 武汉思普崚技术有限公司 Method and system for user network behavior portrayal
CN108920717B (en) * 2018-07-27 2020-03-13 百度在线网络技术（北京）有限公司 Method and device for displaying information
CN109168044B (en) * 2018-10-11 2021-07-23 北京奇艺世纪科技有限公司 Method and device for determining video characteristics
CN109710836A (en) * 2018-11-29 2019-05-03 国政通科技有限公司 A kind of big data intelligent recommendation system and method based on star fan trade council
CN109710890B (en) * 2018-12-20 2023-06-09 四川新网银行股份有限公司 Method and system for identifying false material in real time based on constructed behavior portrait model
CN109934629A (en) * 2019-03-12 2019-06-25 重庆金窝窝网络科技有限公司 A kind of information-pushing method and device
CN110598016B (en) * 2019-09-11 2021-08-17 腾讯科技（深圳）有限公司 Method, device, equipment and medium for recommending multimedia information
CN112825076B (en) * 2019-11-20 2024-03-01 北京搜狗科技发展有限公司 Information recommendation method and device and electronic equipment
CN111915366B (en) * 2020-07-20 2024-01-12 上海燕汐软件信息科技有限公司 User portrait construction method, device, computer equipment and storage medium
CN112383545B (en) * 2020-11-13 2023-07-28 西安热工研究院有限公司 Anti-crawler system, device and deployment method applicable to electric SCADA (supervisory control and data acquisition) system
CN112380418B (en) * 2020-12-31 2021-07-16 金服软件(广州)有限公司 Data processing method and system based on web crawler and cloud platform

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2945113A1 (en) * 2014-05-14 2015-11-18 Cisco Technology, Inc. Audience segmentation using machine-learning
CN105550269A (en) * 2015-12-10 2016-05-04 复旦大学 Product comment analyzing method and system with learning supervising function
CN105718579A (en) * 2016-01-22 2016-06-29 浙江大学 Information push method based on internet-surfing log mining and user activity recognition

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2945113A1 (en) * 2014-05-14 2015-11-18 Cisco Technology, Inc. Audience segmentation using machine-learning
CN105550269A (en) * 2015-12-10 2016-05-04 复旦大学 Product comment analyzing method and system with learning supervising function
CN105718579A (en) * 2016-01-22 2016-06-29 浙江大学 Information push method based on internet-surfing log mining and user activity recognition

Also Published As

Publication number Publication date
CN106599022A (en) 2017-04-26

Similar Documents

Publication Publication Date Title
CN106599022B (en) User portrait forming method based on user access data
CN109492157B (en) News recommendation method and theme characterization method based on RNN and attention mechanism
Zhang et al. Hashtag Recommendation for Multimodal Microblog Using Co-Attention Network.
Venugopalan et al. Exploring sentiment analysis on twitter data
US20160170982A1 (en) Method and System for Joint Representations of Related Concepts
US7987417B2 (en) System and method for detecting a web page template
WO2015191652A1 (en) Modeling interestingness with deep neural networks
Xu et al. Web content mining
CN111309936A (en) Method for constructing portrait of movie user
US20140229486A1 (en) Method and apparatus for unsupervised learning of multi-resolution user profile from text analysis
CN108959323B (en) Video classification method and device
Wu et al. News filtering and summarization on the web
CN107506472B (en) Method for classifying browsed webpages of students
CN112989208B (en) Information recommendation method and device, electronic equipment and storage medium
CN111625715A (en) Information extraction method and device, electronic equipment and storage medium
Liu et al. Correlation identification in multimodal weibo via back propagation neural network with genetic algorithm
CN116362811A (en) Automatic advertisement delivery management system based on big data
CN115659008A (en) Information pushing system and method for big data information feedback, electronic device and medium
Baishya et al. SAFER: sentiment analysis-based fake review detection in e-commerce using deep learning
Fouad et al. Adaptive Visual Sentiment Prediction Model Based on Event Concepts and Object Detection Techniques in Social Media
Jeevanandam Jotheeswaran Sentiment analysis: A survey of current research and techniques
Panchal et al. The social hashtag recommendation for image and video using deep learning approach
Zhong et al. Hignet: hierarchical and interactive gate networks for item recommendation
CN113779387A (en) Industry recommendation method and system based on knowledge graph
CN112417858A (en) Entity weight scoring method, system, electronic equipment 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
CB03 Change of inventor or designer information

Inventor after: Nie Lin

Inventor after: Lin Jing

Inventor after: Wang Qing

Inventor after: Luo Siwei

Inventor before: Luo Siwei

Inventor before: Lin Jing

Inventor before: Wang Qing

Inventor before: Nie Lin

CB03 Change of inventor or designer information
GR01 Patent grant
GR01 Patent grant
EE01 Entry into force of recordation of patent licensing contract

Application publication date: 20170426

Assignee: GUANGDONG TECSUN TECHNOLOGY Co.,Ltd.

Assignor: SUN YAT-SEN University

Contract record no.: X2023980054810

Denomination of invention: A User Profile Formation Method Based on User Access Data

Granted publication date: 20191210

License type: Common License

Record date: 20240102

Application publication date: 20170426

Assignee: Guangzhou Quying Information Technology Co.,Ltd.

Assignor: SUN YAT-SEN University

Contract record no.: X2023980054796

Denomination of invention: A User Profile Formation Method Based on User Access Data

Granted publication date: 20191210

License type: Common License

Record date: 20240102

Application publication date: 20170426

Assignee: SHENDAYUN NETWORK (SHENZHEN) Co.,Ltd.

Assignor: SUN YAT-SEN University

Contract record no.: X2023980054646

Denomination of invention: A User Profile Formation Method Based on User Access Data

Granted publication date: 20191210

License type: Common License

Record date: 20231229

EE01 Entry into force of recordation of patent licensing contract
EE01 Entry into force of recordation of patent licensing contract

Application publication date: 20170426

Assignee: Guangzhou Ainuo Technology Co.,Ltd.

Assignor: SUN YAT-SEN University

Contract record no.: X2024980001983

Denomination of invention: A User Profile Formation Method Based on User Access Data

Granted publication date: 20191210

License type: Common License

Record date: 20240205

Application publication date: 20170426

Assignee: Guangzhou Ruijinyuan Food Technology Co.,Ltd.

Assignor: SUN YAT-SEN University

Contract record no.: X2024980001982

Denomination of invention: A User Profile Formation Method Based on User Access Data

Granted publication date: 20191210

License type: Common License

Record date: 20240205

Application publication date: 20170426

Assignee: Guangzhou Liren Digital Technology Co.,Ltd.

Assignor: SUN YAT-SEN University

Contract record no.: X2024980001991

Denomination of invention: A User Profile Formation Method Based on User Access Data

Granted publication date: 20191210

License type: Common License

Record date: 20240205

Application publication date: 20170426

Assignee: Spectrum Blue Cloud (Guangzhou) Digital Technology Co.,Ltd.

Assignor: SUN YAT-SEN University

Contract record no.: X2024980001990

Denomination of invention: A User Profile Formation Method Based on User Access Data

Granted publication date: 20191210

License type: Common License

Record date: 20240205

Application publication date: 20170426

Assignee: Lingjing Information Technology (Guangzhou) Co.,Ltd.

Assignor: SUN YAT-SEN University

Contract record no.: X2024980001986

Denomination of invention: A User Profile Formation Method Based on User Access Data

Granted publication date: 20191210

License type: Common License

Record date: 20240205

EE01 Entry into force of recordation of patent licensing contract
EE01 Entry into force of recordation of patent licensing contract

Application publication date: 20170426

Assignee: Guangzhou Love Time Information Technology Co.,Ltd.

Assignor: SUN YAT-SEN University

Contract record no.: X2024980002610

Denomination of invention: A User Profile Formation Method Based on User Access Data

Granted publication date: 20191210

License type: Common License

Record date: 20240307

Application publication date: 20170426

Assignee: Zhongyuan Technology (Guangdong) Co.,Ltd.

Assignor: SUN YAT-SEN University

Contract record no.: X2024980002582

Denomination of invention: A User Profile Formation Method Based on User Access Data

Granted publication date: 20191210

License type: Common License

Record date: 20240307

EE01 Entry into force of recordation of patent licensing contract