Effects of relevant contextual features in the performance of a restaurant recommender system
0 likes634 views
The document describes experiments conducted to evaluate the effects of relevant contextual features on the performance of a restaurant recommender system called Surfeous. Key findings include: - Using a reduced subset of attributes (hours, days, accepts, cuisine) performed as well or better than using all attributes, indicating feature selection can improve efficiency. - For recall, subsets generally outperformed a context-free approach, suggesting contextual attributes enrich recommendations. - Fusion achieved similar precision and NDCG as the context-free approach, while rules alone provided lower performance.
![Surfeous: approaches
Social Contextual
[Tso-Sutter et al., 2008]
items
tags
Semantic web
users
items user tags
users R + R Tu
+
item user-based CF Semantic Web Rule Language
R Ti
tags item-based CF (SWRL)
15 / 28](https://image.slidesharecdn.com/oct11-111210181126-phpapp02/85/Effects-of-relevant-contextual-features-in-the-performance-of-a-restaurant-recommender-system-15-320.jpg)
![Feature selection [Guyon & Elisseeff, 2003, Yu et al., 2004]
Generalities Procedures
Machine learning.
Predictive performance. Original
set
Subset
Generation
Subset Subset
Evaluation
Storage requirements. Goodness
of subset
No Yes Result
Stopping
Model understanding. Criterion Validation
Data visualization.
It looks for
the minimum subset of attributes such that the resulting
probability distribution of the data classes is as close as
possible to the original distribution.
19 / 28](https://image.slidesharecdn.com/oct11-111210181126-phpapp02/85/Effects-of-relevant-contextual-features-in-the-performance-of-a-restaurant-recommender-system-19-320.jpg)
![Algorithm LVF (Las Vegas Filter) [Liu & Setiono, 1996]
Input: maximum number of iterations (Max), dataset (D),
number of attributes (N), allowable inconsistency rate (粒)
Output: sets of M features satisfying the inconsistency crite-
rion (Solutions)
Solutions =
Cbest = N
for i = 1 to Max do
S = randomSet(seed); C = numOfFeatures(S)
if C < Cbest then
if InconCheck(S,D) < 粒 then
Sbest = S; Cbest = C
Solutions = S
end if
else if C = Cbest and InconCheck(S,D) < 粒 then
append(Solutions, S)
end if
end for 20 / 28](https://image.slidesharecdn.com/oct11-111210181126-phpapp02/85/Effects-of-relevant-contextual-features-in-the-performance-of-a-restaurant-recommender-system-20-320.jpg)
![Experiments
Data description Attribute selection
111 users. Service contextual model.
237 restaurants. Input: 5,802 instances.
1,251 ratings. Instance: vector of 23
Rating values: 0,1,2. attributes, class=rating.
Rating average: 11.2 Consistency selector
ratings per user. algorithm.
Best-鍖rst search.
65
60
55
Weka [Hall et al., 2009].
50
45
Restaurants
40
35
30
25
20
15
Minimum attribute subset
10
5
0 cuisine, hours, days, accepts,
0 5 10 15 20 25 30 35
Ratings
address (i.e.,78.26% from the
whole set).
22 / 28](https://image.slidesharecdn.com/oct11-111210181126-phpapp02/85/Effects-of-relevant-contextual-features-in-the-performance-of-a-restaurant-recommender-system-22-320.jpg)
Effects of relevant contextual features in the performance of a restaurant recommender system
- 1. Effects of relevant contextual features in the performance of a restaurant recommender system 卒 Blanca Vargas-Govea, Gabriel Gonzalez-Serna, Rafael Ponce-Medell卒n 脹 cenidet - Computer Science Department blanca.vargas@cenidet.edu.mx CARS-2011, October 23, 2011
- 2. Outline 1 Motivation 2 Surfeous-the test bed 3 Feature selection 4 Experiments 5 Conclusions 2 / 28
- 3. Context information / personalization 3 / 28
- 4. Looking for a restaurant? 4 / 28
- 5. Unexpected conditions? 5 / 28
- 6. A better option 6 / 28
- 7. How much context is useful? 7 / 28
- 8. How much context is useful? 8 / 28
- 9. How much context is useful? 9 / 28
- 10. How much context is useful? 10 / 28
- 11. How much context is useful? 11 / 28
- 12. How much context is useful? 12 / 28
- 13. A huge amount of data can be intrusive. A lack of information can lead the system to generate poor recommendations. Approach: attribute selection, semantic models. 13 / 28
- 14. Prototype: Surfeous hungry age: 25 Social likes vegan Context credit card location delicious entrepreneur ugh chinese sunny rainy slow indoor outdoor yum! awful superb noisy 14 / 28
- 15. Surfeous: approaches Social Contextual [Tso-Sutter et al., 2008] items tags Semantic web users items user tags users R + R Tu + item user-based CF Semantic Web Rule Language R Ti tags item-based CF (SWRL) 15 / 28
- 16. Contextual models service (23) environment (2) user (21) 16 / 28
- 17. Rules and relations: examples user - service pro鍖le person(X ) hasOccupation(X , student) restaurant(R) hasCost(R, low) select(X , R) user - environment pro鍖le person(X ) isJapanese(X , true) queryPlace(X , USA) restaurant(R) isVeryClose(R, true) select(X , R) environment - service pro鍖le currentWeather(today, rainy) restaurant(R) space(R, closed) select(R) Relations likesFood(X , Y ) X : person, Y : cuisine-type currentWeather(X , Y ) X : query, Y : weather space(X , Y ) X : restaurant, Y : {closed, open} 17 / 28
- 18. Generating recommendations 1 2 3 ambiance city cuisine space accepts latitude Surfeous gets the user Relations are created location and searches for An ontology is created from the attributes of the the closer restaurants in execution time restaurant profile 4 5 Results are 6 Fusion ranked based context-free context Person(?x) ^ hasAge(?x, ?y) ^ Ranking 1. ---------- on the number swrlb:greaterThanOrEqual(?y, 12) ^ 2. ---------- of context only-social only-rules swrlb:lessThanOrEqual... 3. ---------- 0% 100% rules that hold SWRL is applied to match ... The social results are for each n. ---------- added the context models user query 18 / 28
- 19. Feature selection [Guyon & Elisseeff, 2003, Yu et al., 2004] Generalities Procedures Machine learning. Predictive performance. Original set Subset Generation Subset Subset Evaluation Storage requirements. Goodness of subset No Yes Result Stopping Model understanding. Criterion Validation Data visualization. It looks for the minimum subset of attributes such that the resulting probability distribution of the data classes is as close as possible to the original distribution. 19 / 28
- 20. Algorithm LVF (Las Vegas Filter) [Liu & Setiono, 1996] Input: maximum number of iterations (Max), dataset (D), number of attributes (N), allowable inconsistency rate (粒) Output: sets of M features satisfying the inconsistency crite- rion (Solutions) Solutions = Cbest = N for i = 1 to Max do S = randomSet(seed); C = numOfFeatures(S) if C < Cbest then if InconCheck(S,D) < 粒 then Sbest = S; Cbest = C Solutions = S end if else if C = Cbest and InconCheck(S,D) < 粒 then append(Solutions, S) end if end for 20 / 28
- 21. Toy example space price franchise smoking RatingA RatingB 1 i low n y 0 0 2 i low n y 1 0 3 i low n y 2 0 4 i low n y 1 1 5 i high n n 0 1 6 i high n n 1 1 7 i high n n 2 1 8 o high y n 1 1 9 o low n n 1 1 10 o low n y 2 2 subset A subset B matching instances: 1, 2, 3, 4 matching instances: 1, 2, 3, 4 n = 4, classes = 0,1,2,1 largest = 1 (2 n = 4, classes = 0,0,0,1 largest = 0 (3 instances) instances) Inconsistency count = 4 - 2 = 2 Inconsistency count = 4 - 3 = 1 matching instances: 5, 6, 7 matching instances: 5, 6, 7 n = 3, classes = 0,1,2 largest = 1 (1 n = 3, classes = 1,1,1 largest = 1 (3 instances) instances) Inconsistency count = 3 - 1 = 2 Inconsistency count = 3 - 3 = 0 Inconsistency rate = (2+2)/10 = 4/10 = 0.4 Inconsistency rate = (1+0)/10 = 1/10 = 0.1 21 / 28
- 22. Experiments Data description Attribute selection 111 users. Service contextual model. 237 restaurants. Input: 5,802 instances. 1,251 ratings. Instance: vector of 23 Rating values: 0,1,2. attributes, class=rating. Rating average: 11.2 Consistency selector ratings per user. algorithm. Best-鍖rst search. 65 60 55 Weka [Hall et al., 2009]. 50 45 Restaurants 40 35 30 25 20 15 Minimum attribute subset 10 5 0 cuisine, hours, days, accepts, 0 5 10 15 20 25 30 35 Ratings address (i.e.,78.26% from the whole set). 22 / 28
- 23. Tests with Surfeous Purposes Experimental setup to identify relevant Leave one out. contextual attributes. Seven subsets: All (23), B to show that with the (5), C-G (4). minimum attribute subset, the predictive performance 10 executions for each is at least the same as with subset. the whole attribute set, and Baseline: context-free, to analyze the effects of fusion (average of the relevant contextual intervals between 0.1 and attributes. 0.9) and context (only rules). 23 / 28
- 24. Results: precision/recall/NDCG 0.09 0.35 0.08 0.30 0.07 0.25 0.06 type type Precision 0.05 0.20 Recall context.free context.free 0.04 0.15 fusion fusion 0.03 context 0.10 context 0.02 0.05 0.01 All B C D E F G All B C D E F G subset subset 0.55 0.50 0.45 0.40 0.35 type 0.30 NDCG context.free 0.25 fusion 0.20 0.15 context 0.10 0.05 All B C D E F G subset All (23), B (cuisine, hours, days, accepts, address), C (cuisine, hours, days), D (hours, days, accepts, address), E(cuisine, days, accepts, address), F (cuisine, hours, accepts, address), G (cuisine, hours, days, accepts) 24 / 28
- 25. Precision Recall NDCG Fusion D C D Rules F C G Relevant attributes: hours, days, accepts, cuisine. 25 / 28
- 26. Precision Recall NDCG Fusion D C D Rules F C G Relevant attributes: hours, days, accepts, cuisine. For recall, the majority of the subsets outperformed the context-free performance. 25 / 28
- 27. Precision Recall NDCG Fusion D C D Rules F C G Relevant attributes: hours, days, accepts, cuisine. For recall, the majority of the subsets outperformed the context-free performance. For precision and NDCG, fusion obtained similar performance to the context-free approach. 25 / 28
- 28. Precision Recall NDCG Fusion D C D Rules F C G Relevant attributes: hours, days, accepts, cuisine. For recall, the majority of the subsets outperformed the context-free performance. For precision and NDCG, fusion obtained similar performance to the context-free approach. Expected items appear in the top-5 list. 25 / 28
- 29. Precision Recall NDCG Fusion D C D Rules F C G Relevant attributes: hours, days, accepts, cuisine. For recall, the majority of the subsets outperformed the context-free performance. For precision and NDCG, fusion obtained similar performance to the context-free approach. Expected items appear in the top-5 list. Results suggest that the restaurant opening times and its type of payment are likely to be the most important factors to make a choice. 25 / 28
- 30. Precision Recall NDCG Fusion D C D Rules F C G Relevant attributes: hours, days, accepts, cuisine. For recall, the majority of the subsets outperformed the context-free performance. For precision and NDCG, fusion obtained similar performance to the context-free approach. Expected items appear in the top-5 list. Results suggest that the restaurant opening times and its type of payment are likely to be the most important factors to make a choice. Although the performance achieved by the semantic rules is low, they provide the social approach with features that enriches the decision process (recall). A deep analysis of the set of rules is needed. 25 / 28
- 31. Conclusions and future work By using a reduced subset of attributes, the systems performance was not degraded. Moreover, in the fusion approach it has been improved. Feature selection techniques can contribute to improve the ef鍖ciency of a contextual recommender system. Identi鍖cation of relevant contextual features facilitates a better understanding of the decision criteria of users. As part of our future work, we are extending the approach to the three contextual models. 26 / 28
- 32. Effects of relevant contextual features in the performance of a restaurant recommender system Blanca Vargas-Govea blanca.vargas@cenidet.edu.mx CARS-2011, October 23, 2011 27 / 28
- 33. Creative Commons licensed images s04 - Outdoor restaurant s05 - Gray umbrella s06 - Indoor restaurant s14 - Sunny s14 - Red umbrella s14 - Indoor restaurant s14 - Outdoor restaurant s14 - Crowd s14 - Chinese restaurant s14 - Persian girl s18 - Earth s18 - Waf鍖e 28 / 28
- 34. Guyon, I. & Elisseeff, A. (2003). An introduction to variable and feature selection. Journal of Machine Learning Research, 3, 11571182. Hall, M., Frank, E., Holmes, G., Pfahringer, B., Reutemann, P., & Witten, I. H. (2009). The WEKA data mining software: an update. SIGKDD Explorations Newsletter, 11, 1018. Liu, H. & Setiono, R. (1996). A probabilistic approach to feature selection - a 鍖lter solution. In 13th International Conference on Machine Learning (pp. 319327). Tso-Sutter, K. H. L., Marinho, L. B., & Schmidt-Thieme, L. (2008). Tag-aware recommender systems by fusion of collaborative 鍖ltering algorithms. In Proceedings of the 2008 ACM symposium on Applied computing (pp. 19951999). New York, USA. Yu, L., Liu, H., & Guyon, I. (2004). Ef鍖cient feature selection via analysis of relevance and redundancy. Journal of Machine Learning Research, 5, 12051224. 28 / 28