Abstract | ||
---|---|---|
•A Conformal Recommender System (CRS) is developed that blends the concept of Conformal Prediction(CP) with Recommender System. It may be noted that there has not been any earlier attempt in this direction and it is certainly useful for the user to know the level of confidence of any recommendation.•We propose a novel way of defining nonconformity measure for recommender systems by taking into consideration the precedence probability among items.•We show empirically that all desired properties of conformal prediction such as validity and efficiency are observed by the proposed measure.•We show through extensive experimentation that in this process we are in a position to have a better recommender system in terms of assigning confidence values to the recommended items and also in terms of more accurate recommendation. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1016/j.ins.2017.04.005 | Information Sciences |
Keywords | DocType | Volume |
Recommender systems,Conformal prediction,Confidence | Journal | 405 |
ISSN | Citations | PageRank |
0020-0255 | 1 | 0.35 |
References | Authors | |
0 | 5 |
Name | Order | Citations | PageRank |
---|---|---|---|
Venkateswara Rao Kagita | 1 | 59 | 8.13 |
Arun K. Pujari | 2 | 420 | 48.20 |
Vineet Padmanabhan | 3 | 216 | 25.90 |
Sandeep Kumar Sahu | 4 | 19 | 2.63 |
Vikas Kumar 0003 | 5 | 25 | 4.76 |