Personalized TV Recommendation with Mixture Probabilistic Matrix Factorization

Size: px

Start display at page:

Download "Personalized TV Recommendation with Mixture Probabilistic Matrix Factorization"

Reginald Lynch
5 years ago
Views:

1 Personalized TV Recommendation with Mixture Probabilistic Matrix Factorization Huayu Li, Hengshu Zhu #, Yong Ge, Yanjie Fu +,Yuan Ge Computer Science Department, UNC Charlotte # Baidu Research-Big Data Lab + Rutgers University Anhui Polytechnic University 5/1/2015 1

2 Outline Introduction Challenges of TV Recommendation Data Methods Experiments Conclusion 5/1/2015 2

3 Introduction Nowadays, smart TV is very prevalent 5/1/2015 3

4 Introduction However, which TV program should we watch? 5/1/2015 4

5 Introduction TV Recommender System is very important! However, which TV program should we watch? 5/1/2015 5

6 Outline Introduction Challenges of TV Recommendation Data Methods Experiments Conclusion 5/1/2015 6

7 Television Watching Groups TV Program 7

8 Watching group refers to users who have similar preferences for TV programs in front of a television. Television Watching Groups TV Program 8

9 Challenges of TV Recommendation 1. How to infer the preference for different watching group from such a large number of individual watching records? 2. How to handle the implicit feedbacks of users, e.g. watching time? 5/1/2015 9

10 Outline Introduction Challenges of TV Recommendation Data Methods Experiments Conclusion 5/1/

11 Data 1. Each watching record includes: Television ID Program ID Time Information For example : TV ID Program ID Watching Duration Start Time Total Time 2 ba T00:00:00.000Z Each TV program includes: Title Two types of genres: first level genre and sub-genre 5/1/

12 Data 1. Each watching record includes: Television ID Program ID Time Information For example : # Televisions # TV Programs # Watching Records TV ID Program ID Watching Duration Start Time Total Time 230,196 4,289 14,159,678 2 ba T00:00:00.000Z Each TV program includes: Title Two types of genres: first level genre and sub-genre 5/1/

13 Outline Introduction Challenges of TV Recommendation Data Methods Experiments Conclusion 5/1/

14 Methods Basic Framework Step 1: Discover Watching Groups Step 2: Learn Preference of Television 5/1/

15 Methods Basic Framework Step 1: Discover Watching Groups Step 2: Learn Preference of Television 5/1/

16 Methods Discovery of Watching Groups Television Clustering (K-means) Feature: Watching frequency of TV program Estimating Watching Groups (Markov Clustering) Feature: First-level genre Sub-genre Watching time in a day Week day or weekend 5/1/

17 Methods Discovery of Watching Groups TV Group 1 TV Group 2 17

18 Methods Discovery of Watching Groups TV Group 1 TV Group 2 In each TV group, televisions have similar watching groups. 18

19 Methods Discovery of Watching Groups Television Clustering (K-means) Feature: Watching frequency of TV program Estimating Watching Groups (Markov Clustering) Feature: First-level genre Sub-genre Watching time in a day Week day or weekend 5/1/

20 Methods Discovery of Watching Groups Television Clustering (K-means) Feature: Watching frequency of TV program Estimating Watching Groups (Markov Clustering) Feature: First-level genre Sub-genre Watching time in a day Week day or weekend 5/1/

21 Methods Discovery of Watching Groups TV Group 1 TV Group 2 TV Groups The hidden watching group number 21

22 Methods mpmf Basic frame work Step 1: Discover Watching Groups Step 2: Learn Preference of Television Mixture Probabilistic Matrix Factorization (mpmf) 5/1/

23 Methods mpmf Assumption: The preferences of a television for TV programs could be decomposed into a mixture preference of the hidden watching groups. Preference of TV Preferences of Watching Groups Mixture 5/1/

24 Methods mpmf Given: The learned number of watching groups for each television group Television Program R = Television K T K Program V 5/1/

25 Methods mpmf Given: The learned number of watching groups for each television group 1. Draw television-specific latent factor from a mixture of Gaussian distribution 2. The mixture number is the number of watching groups 5/1/

26 Methods mpmf 5/1/

27 Methods mpmf Alternating Least Square for the parameter estimation. 5/1/

28 Outline Introduction Challenges of TV Recommendation Data Methods Experiments Conclusion 5/1/

29 Experiments Show an example of clustering Evaluate the proposed method s performance Prediction Accuracy Ranking Accuracy Top-K Recommendation Compare different data conversion methods 5/1/

30 Experiments An Example of clustering An example of clustering: Left is the clustering result, and Right is the corresponding program names and main genres. 5/1/

31 Experiments Prediction Accuracy Rating Conversion Cumulative ratio of watching time to the total time of a program played Baselines PMF mpmf Random # watching group # watching group as 1 # watching group as 3 5/1/

32 Experiments Ranking Accuracy 5/1/

33 Experiments Top-K Recommendation 5/1/

34 Experiments Top-K Recommendation 5/1/

35 Experiments Data Conversion Methods Data Conversion Methods Cumulative Ratios Frequency Binary Confidence Level 5/1/

36 Conclusion Design a two-stage framework Employ clustering to discover watching groups Develop probabilistic model to learn the preference of television for TV program based on Gaussian mixture distributions Evaluate the proposed model in real-world data with various metrics 5/1/

37 Thank you! Question? 5/1/

Personalized TV Recommendation with Mixture Probabilistic Matrix Factorization

Personalized TV Recommendation with Mixture Probabilistic Matrix Factorization Huayu Li Hengshu Zhu Yong Ge Yanjie Fu Yuan Ge ± Abstract With the rapid development of smart TV industry, a large number