Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

KNN and three-way decision-based movie recommendation method

A recommendation method and film technology, applied in the fields of electrical digital data processing, special data processing applications, instruments, etc., can solve the problems of different scoring standards, large scoring gaps, distrust, etc., and achieve accurate prediction and improved recommendation quality. Effect

Inactive Publication Date: 2016-10-12
WUHAN UNIV OF TECH
View PDF2 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The above method has great flaws. For example, there may be the following situations. For example, although two users have the same evaluation on a certain movie, their rating standards may be different, resulting in a large gap between their ratings. In fact, this Two users have the same interest; in addition, for some movies, the similarity of user ratings may not be recommended as the primary factor. For some experts who study movie art, he may be more willing to accept some movies that have the same Only professional people can make recommendations to achieve better results; another important factor is the credibility of users. Now there are many distrustful things happening on the Internet, such as fake ratings and reviews, etc.
Because there are many online movie rental companies that may hire someone to increase the rating of a certain movie in order to increase the ratings of a certain movie.

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • KNN and three-way decision-based movie recommendation method
  • KNN and three-way decision-based movie recommendation method
  • KNN and three-way decision-based movie recommendation method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0068] In the first embodiment, the sample set D uses the Book-Crossing dataset.

[0069] The sample set was collected by Cai-Nicolas Ziegler through the Book-Crossing website. It contains 1,149,780 rating information of 271,379 books from 278,858 users. The stronger the user interest.

[0070] In order to facilitate the experiment, the scoring value is re-calibrated, and the scoring value of 9 and 10 is marked as +1 (recommended sample set), and the score value of 0-8 is marked as -1 (not recommended sample set). This embodiment is performed on the scored data of the first 1000 items in the Book-Crossing dataset. About 35,000 users rated these 1,000 items, and the sub-sample set contains a total of more than 140,000 pieces of rating data.

[0071] Such as figure 2 As shown, the accuracy of recommending the information in the Book-Crossing data set according to the KNN algorithm and the solution provided by the present invention is shown.

Embodiment 2

[0072] In the second embodiment, the sample set D is taken from the MovieLens data set.

[0073] The sample set was collected by the GroupLens research team at the University of Minnesota through the MovieLens website. It contains 1,000,000 ratings from 1 to 5 points for 1,682 movies by 943 users, and each user rated at least 20 movies. Like the Book-Crossing data set, the score value is re-calibrated, and the score value of 4 and 5 is marked as +1 (recommended sample set), and the score value of 1-3 is marked as -1 (not recommended sample set). ). The scoring data of 100, 200, and 300 users are randomly selected from the MovieLens dataset to form three sample sets, which are recorded as TDS100, TDS200, and TDS300.

[0074] Such as figure 2 As shown, the accuracy of recommending the information in the Book-Crossing data set according to the KNN algorithm and the solution provided by the present invention is shown.

[0075] According to the experimental results, the solutio...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a KNN and three-way decision-based movie recommendation method. The method comprises the following steps: receiving a selection condition input by a user; dividing a movie sample set D into a movie sample set XP with a label P and a movie sample set Xn with a label N according to the received selection condition; calculating m nearest neighbor sets and corresponding label sets, in the movie sample set XP, of the ith movie sample x(i) in the movie sample set D as well as n nearest neighbor sets and corresponding label sets, in the movie sample set XN, of the ith movie sample x(i) in the movie sample set D ac to KNN; calculating a posterior probability P(XN / x(i)), in the movie sample sets XP and XN, of the movie sample x(i); and calculating a recommendation value (the formula is as shown in the specification) of the movie sample x(i) according to a movie three-way decision, and determining to classify the movie sample x(i) according to the calculated recommendation value (the formula is as shown in the specification).

Description

technical field [0001] The invention relates to the technical field of information recommendation, in particular to a movie recommendation method based on KNN and three-way decision-making. Background technique [0002] The personalized recommendation system is also referred to as the recommendation system. Its basic working principle is to analyze the user's habits and hobbies based on the user's behavior data, such as the user's quantitative rating data or non-quantitative comment data on a certain movie, and then provide targeted recommendations to the user. Users recommend content that they may be interested in, which is currently widely used in many Internet fields such as e-commerce, television, film and video, music, social networking, and other fields. Amazon is currently a very famous e-commerce website, and it is also the earliest practitioner and promoter of personalized recommendation system. Its main applications are personalized recommendation list and related ...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06F17/30
CPCG06F16/9535
Inventor 熊盛武郑文博段鹏飞于笑寒周姜炜
Owner WUHAN UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com