LU502892B1 - User Label Weight Evaluation Method Based on Fuzzy Theory - Google Patents
User Label Weight Evaluation Method Based on Fuzzy Theory Download PDFInfo
- Publication number
- LU502892B1 LU502892B1 LU502892A LU502892A LU502892B1 LU 502892 B1 LU502892 B1 LU 502892B1 LU 502892 A LU502892 A LU 502892A LU 502892 A LU502892 A LU 502892A LU 502892 B1 LU502892 B1 LU 502892B1
- Authority
- LU
- Luxembourg
- Prior art keywords
- label
- labels
- user
- weight
- weights
- Prior art date
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
- G06F16/90—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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/3331—Query processing
- G06F16/334—Query execution
- G06F16/3346—Query execution using probabilistic model
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/12—Computing arrangements based on biological models using genetic models
- G06N3/126—Evolutionary algorithms, e.g. genetic algorithms or genetic programming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/02—Computing arrangements based on specific mathematical models using fuzzy logic
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2216/00—Indexing scheme relating to additional aspects of information retrieval not explicitly covered by G06F16/00 and subgroups
- G06F2216/03—Data mining
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Software Systems (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Algebra (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Probability & Statistics with Applications (AREA)
- Computational Mathematics (AREA)
- Automation & Control Theory (AREA)
- Fuzzy Systems (AREA)
- Physiology (AREA)
- Genetics & Genomics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention provides a user label weight evaluation method based on fuzzy theory, including the following steps: 1), user label analysis and weight initialization: classifying user labels into a basic label, a behavior label, and a social label, analyzing the importance of the labels, and then initializing the weights of all the sub-labels of the three labels; 2) determining weight variation intervals of all the sub-labels: grading the sub-labels under each label, using a fuzzy set to generate a corresponding membership function MDF for each user label weight, and determining the weight variation intervals of all the labels according to MDF ; and 3), designing a particle swarm optimization algorithm to optimize the user label weights; initializing an attribute weight of each particle according to the weight variation interval of each label, and optimizing the user label weights by means of the particle swarm optimization algorithm. The application of the technical solution can more comprehensively consider the user labels.
Description
User Label Weight Evaluation Method Based on Fuzzy Theory
The present invention relates to the technical field of network data mining, and more particularly to a user label weight evaluation method based on fuzzy theory.
With the popularization of mobile devices, the iterative update of wireless technologies, and the rapid development of a mobile social network, more and more users participate in the mobile social network to communicate and share information. With the sharp increase of user quantity, data in the network also grows nonlinearly. Massive data contains huge values. However, user information is complex, and has the situations of information missing and false information. A user configuration file refers to a labeled user model abstracted from the information such as a basic attribute of a user, a user preference, a living habit, a user behavior and the like. Each label and each label weight are a vector of a user. And a user can be understood as a sum of multiple vectors (labels) in a super-dimensional space. The user described by the data is finally recognized by a computer, on the basis of which the user file is applied. The determination of the label weights has a huge influence on subsequent user profile based recommendation and precise marketing. Existing label weight algorithms are mainly on the basis of the TF-IDF algorithm proposed by Sparck.
However, the current algorithms do not consider user labels comprehensively.
In view of the existing status, the objective of the present invention is to provide a user label weight evaluation method based on fuzzy theory, which gives more comprehensive consideration to user labels.
To achieve the above objective, the present invention adopts the following technical solution: a user label weight evaluation method based on fuzzy theory, including the following steps: step S1, user label analysis and weight initialization: classifying user labels into a basic label, a behavior label, and a social label, analyzing the importance of the labels, and then initializing the weights of all the sub-labels of the basic label, the behavior label, and the social label; step S2: determining weight variation intervals of all the sub-labels: grading the sub-labels under each label, using a fuzzy set to generate a corresponding membership function MDF for each user label weight, and determining the weight variation intervals of all the labels according to the Lu502892 membership function MDF; and step S3, designing a particle swarm optimization algorithm to optimize the user label weights: randomly initializing attribute weights of a group of particles according to the weight variation intervals of all the labels, and optimizing the user label weights by means of the particle swarm optimization algorithm.
In a preferred embodiment, the step S2 specifically includes: step S201, grading the sub-labels under each user label into "low", "medium", and "high" according to assigned initial weights; step S202, using a Gaussian formula to generate a fuzzy set for the weights of the three grades of sub-labels under the label, and generating a membership function WDF corresponding to the label according to the fuzzy set, wherein a variance of the membership function MDF is determined by an interval range formed by initial weight values; and step S203, obtaining the weight variation intervals of the three grades of sub-labels under each label according to the corresponding membership function MDF and a maximum MD principle.
In a preferred embodiment, the step S3 designing a particle swarm optimization algorithm to optimize the user label weights specifically includes: step S301, generating a group of particles, and treating all the user labels as attributes of the particles; step S302, randomly initializing the user label weights of all the particles according to the weight variation intervals of all the labels; and step S303, optimizing the user label weights by means of the particle swarm optimization algorithm, setting an optimization convergence condition that an optimal particle does not change any more or has reached a maximum number of iterations, and solving the optimal label weight, namely the label weight of the optimal particle.
Compared with the prior art, the present invention has the following beneficial effects:
Compared with TF-IDF, the present invention converts the problem of evaluating the user label weights into the problem of seeking optimal solutions, designs a membership function to obtain a fuzzy boundary of all the user label weights, and uses the genetic algorithm to obtain the optimal solution of each user label weight. The present invention divides different types of user labels into a basic label, a network label, and a behavior label, which is more coincident with the background pf)502892 mobile social network.
The present invention adopts the above technical solution, and provides a user label weight evaluation method based on fuzzy theory in the field of network data mining. In the present invention, the importance of the user labels is analyzed, and the weights of the labels are initialized; the sub-labels of each label are graded, and the fuzzy theory is used to generate a membership function MDF of each label; the weight variation interval of each grade of sub-labels is calculated according to the membership function MDF; and a particle swarm optimization algorithm is designed to optimize the user label weights.
Compared with the other methods, the present invention considers the user labels more comprehensively, converts the problem of evaluating the user label weights into the problem of seeking optimal solutions, designs a membership function to obtain a fuzzy boundary of all the user label weights, and uses the particle swarm optimization to obtain the optimal solution of each user label weight.
Fig. 1 is a flow chart of a user label weight evaluation method based on fuzzy theory according to a preferred embodiment of the present invention; and
Fig. 2 is a view of a membership function MDF according to a preferred embodiment of the present invention.
The present invention will be further described below with reference to the accompanying drawings and embodiments.
It should be pointed out that the detailed descriptions below are illustrative, and are intended to provide further description for the present application. Unless otherwise specified, all the technical and scientific terms used in the text have the same meanings as that a person skilled in the art generally understands.
It should be noted that the terms used herein are only used to describe specific embodiments, but not intended to limit the exemplary embodiments of the present application. As used herein, unless otherwise specified in the context, a singular form is also intended to include a plural form.
In addition, it should be understood that the terms "include" and/or "comprise" used in the specification indicate the existence of a feature, a step, an operation, a device, an assembly, and/or a combination thereof.
The present invention provides a user label weight evaluation method based on fuzzy theory 502892
In order to achieve the objective, firstly, user labels are analyzed and weights are initialized; then, a membership function MDF is designed to determine weight variation intervals of all the sub-labels; and finally, a particle swarm optimization algorithm is used to optimize the user label weights according to the initialized user label weights, user label scores, and the weight variation intervals.
With reference to figures 1-2, the present invention discloses a user label weight evaluation method based on fuzzy theory, including the following steps:
Step S1, user label analysis and weight initialization: classifying user labels into a basic label, a behavior label, and a social label, analyzing the importance of the labels, and then initializing the weights of all the sub-labels of the three labels; step S2: determining weight variation intervals of all the sub-labels, as shown in figure 2: grading the sub-labels under each label, using a fuzzy set to generate a corresponding membership function MDF for each user label weight, and determining the weight variation intervals of all the labels according to MDF; and step S3, designing a genetic algorithm to optimize the user label weights: calculating a fitness function according to all the user label weights, using the fitness function to design a genetic algorithm GA, and optimizing the user label weights by means of the GA.
The step S2 specifically includes: step S201, grading the sub-labels under each user label into "low", "medium", and "high" according to assigned initial weights; step S202, using a Gaussian formula to generate a fuzzy set for the weights of the three grades of sub-labels under the label, and generating a membership function MDF corresponding to the label according to the fuzzy set, wherein a variance of the MDF is determined by an interval range formed by initial weight values; and step S203, obtaining the weight variation intervals of the three grades of sub-labels under each label according to the corresponding membership function MDF and a maximum MD principle.
The step S3 designing a particle swarm optimization algorithm to optimize the user label weights specifically includes: step S301, generating a group of particles, and treating all the user labels as attributes of the particles;
step S302, randomly initializing the user label weights of all the particles according to thejs02892 weight variation intervals of all the labels; and step S303, optimizing the user label weights by means of the particle swarm optimization algorithm, setting an optimization convergence condition that an optimal particle does not change 5 any more or has reached a maximum number of iterations, and solving the optimal label weight, namely the label weight of the optimal particle.
The present invention adopts the above technical solution, and provides a user label weight evaluation method based on fuzzy theory in the field of network data mining. In the present invention, the importance of the user labels is analyzed, and the weights of the labels are initialized, the sub-labels of each label are graded, and the fuzzy theory is used to generate a membership function MDF of each label; the weight variation interval of each grade of sub-labels is calculated according to the membership function MDF; and a particle swarm optimization algorithm is designed to optimize the user label weights.
Compared with the other methods, the present invention considers the user labels more comprehensively, converts the problem of evaluating the user label weights into the problem of seeking optimal solutions, designs a membership function to obtain a fuzzy boundary of all the user label weights, and uses the particle swarm optimization to obtain the optimal solution of each user label weight.
The descriptions above are only the preferred embodiments of the present invention, but are not intended to limit the patent scope of the present invention. Any equivalent substitutions made by using the specification and the drawings of the present invention, or direct or indirect applications in related technical fields are all concluded in the scope of protection of the present invention for the same reason.
Claims (3)
1. A user label weight evaluation method based on fuzzy theory, comprising the following steps: step S1, user label analysis and weight initialization: classifying user labels into a basic label, a behavior label, and a social label, analyzing the importance of the labels, and then initializing the weights of all the sub-labels of the basic label, the behavior label, and the social label; step S2: determining weight variation intervals of all the sub-labels: grading the sub-labels under each label, using a fuzzy set to generate a corresponding membership function MDF for each user label weight, and determining the weight variation intervals of all the labels according to the membership function MDF; and step S3, designing a particle swarm optimization algorithm to optimize the user label weights: randomly initializing attribute weights of a group of particles according to the weight variation intervals of all the labels, and optimizing the user label weights by means of the particle swarm optimization algorithm.
2. The user label weight evaluation method based on fuzzy theory according to claim 1, wherein the step S2 specifically comprises: step S201, grading the sub-labels under each user label into "low", "medium", and "high" according to assigned initial weights; step S202, using a Gaussian formula to generate a fuzzy set for the weights of the three grades of sub-labels under the label, and generating a membership function MDF corresponding to the label according to the fuzzy set, wherein a variance of the membership function MDF is determined by an interval range formed by initial weight values; and step S203, obtaining the weight variation intervals of the three grades of sub-labels under each label according to the corresponding membership function MDF and a maximum MD principle.
3. The user label weight evaluation method based on fuzzy theory according to claim 1, wherein the step S3 designing a particle swarm optimization algorithm to optimize the user label weights specifically comprises: step S301, generating a group of particles, and treating all the user labels as attributes of the particles; step S302, randomly initializing the user label weights of all the particles according to the weight variation intervals of all the labels; and step S303, optimizing the user label weights by means of the particle swarm optimization | y502892 algorithm, setting an optimization convergence condition that an optimal particle does not change any more or has reached a maximum number of iterations, and solving the optimal label weight, namely the label weight of the optimal particle.
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211037501.8A CN115357777B (en) | 2022-08-26 | 2022-08-26 | Fuzzy theory-based user tag weight evaluation method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| LU502892B1 true LU502892B1 (en) | 2023-04-18 |
Family
ID=84005349
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| LU502892A LU502892B1 (en) | 2022-08-26 | 2022-10-12 | User Label Weight Evaluation Method Based on Fuzzy Theory |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN115357777B (en) |
| LU (1) | LU502892B1 (en) |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6701312B2 (en) * | 2001-09-12 | 2004-03-02 | Science Applications International Corporation | Data ranking with a Lorentzian fuzzy score |
| US8718534B2 (en) * | 2011-08-22 | 2014-05-06 | Xerox Corporation | System for co-clustering of student assessment data |
| CN104933862A (en) * | 2015-05-26 | 2015-09-23 | 大连理工大学 | Urban traffic jam intelligent combination prediction method based on track of floating vehicle |
| CN106096723B (en) * | 2016-05-27 | 2018-10-30 | 北京航空航天大学 | It is a kind of that complex industrial properties of product appraisal procedure is used for based on hybrid neural networks algorithm |
| CN109190675A (en) * | 2018-08-06 | 2019-01-11 | 浪潮电子信息产业股份有限公司 | A kind of Fuzzy classification and device based on particle swarm optimization algorithm |
| CN109215350B (en) * | 2018-11-02 | 2021-05-25 | 重庆大学 | Short-term traffic state prediction method based on RFID electronic license plate data |
| CN110851272B (en) * | 2019-10-30 | 2022-02-11 | 内蒙古农业大学 | Cloud task scheduling method based on phagocytic particle swarm genetic hybrid algorithm |
| CN111860097B (en) * | 2020-04-16 | 2024-02-20 | 西安电子科技大学 | Abnormal behavior detection method based on fuzzy theory |
-
2022
- 2022-08-26 CN CN202211037501.8A patent/CN115357777B/en active Active
- 2022-10-12 LU LU502892A patent/LU502892B1/en active IP Right Grant
Also Published As
| Publication number | Publication date |
|---|---|
| CN115357777A (en) | 2022-11-18 |
| CN115357777B (en) | 2023-09-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Xin et al. | An adaptive random walk sampling method on dynamic community detection | |
| Yu et al. | A semantic overlapping community detection algorithm based on field sampling | |
| CN106682527A (en) | Data security control method and system based on data classification and grading | |
| CN105893406A (en) | Group user profiling method and system | |
| Xin et al. | An overlapping semantic community detection algorithm base on the ARTs multiple sampling models | |
| CN112508726B (en) | False public opinion identification system based on information spreading characteristics and processing method thereof | |
| Xin et al. | The adaptive dynamic community detection algorithm based on the non-homogeneous random walking | |
| Balakrishna et al. | An efficient incremental clustering based improved K-Medoids for IoT multivariate data cluster analysis | |
| WO2023207013A1 (en) | Graph embedding-based relational graph key personnel analysis method and system | |
| Ghoson | Decision tree induction & clustering techniques in SAS Enterprise Miner, SPSS Clementine, and IBM Intelligent Miner: a comparative analysis | |
| Choi et al. | K-Means and J48 Algorithms to Categorize Student Research Abstracts | |
| Cui et al. | A k-means++ based user classification method for social e-commerce | |
| LU502892B1 (en) | User Label Weight Evaluation Method Based on Fuzzy Theory | |
| Binoti et al. | Estimation of height of Eucalyptus trees with neuroevolution of Augmenting Topologies (NEAT) | |
| CN106874447A (en) | A kind of method for exhibiting data and device | |
| Zhong et al. | Recommendations for mobile apps based on the hits algorithm combined with association rules | |
| Luo et al. | DeepAttr: Inferring demographic attributes via social network embedding | |
| Zhao et al. | Intrusion Detection Based on Improved SOM with Optimized GA. | |
| Zhou et al. | Retrofitting embeddings for unsupervised user identity linkage | |
| Srinivasarao et al. | Introduction to data science: Review, challenges, and opportunities | |
| Ye et al. | Incorporating side information into multivariate Information Bottleneck for generating alternative clusterings | |
| Teimouri et al. | Modeling tree diameters using mixtures of skewed Student’st and related distributions | |
| Kumar et al. | An analysis and literature review of algorithms for frequent itemset mining | |
| Wang et al. | An api recommendation method based on beneficial interaction | |
| Vanarse et al. | MULTILEVEL ASSOCIATION RULE MINING FOR LARGE DATASETS: A |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| FG | Patent granted |
Effective date: 20230418 |