SOCIAL BOOK SEARCH TRACK: ISM@INEX’14 SUGGESTION TASK Ritesh Kumar and Sukomal Pal Department of Computer Science and Engineering, Indian School of Mines Dhanbad, 826004 India {ritesh4rmrvs,sukomalpal}@gmail.com Abstract. This paper describes the work that we did at Indian School of Mines towards Social Book Search Track for INEX 2014. We submit- ted five runs in its Suggestion Task. We investigated individual effect of title, group, mediated query, and narrative fields of the topics in our runs. For all the runs we used language modelling technique with Dirich- let smoothing. The run using only mediated query field was our best. Overall, our performance is not satisfactory. However, as new entrant to the field, our scores are encouraging enough to work for better results in future. Keywords: Book Search, Social Book Search, Language modelling, In- formation Retrieval 1 Introduction With growing numbers of online portals and book catalogues, our current time sees a rapid evolution in the way we acquire, share and use books. In order to en- able users, Social Book Seach Track at INEX [5] provides a relevant experimental platform to investigate techniques of searching and navigating professional meta- data provided by publishers/booksellers and user-generated content from social media [1]. At INEX 2014, they offered two tasks: Suggestion Task and Interactive Task. We participated in the first where we were supposed to recommend books based on user’s request and her personal catalogue data (list of books with rating and tags maintained for the user in the social cataloguing site). We were also provided with a large set of anonymised user profiles from LibraryThing forum members. Each user request is provided in the form of topics containing different fields like title, mediated query, group, narrative and catalogue information. As a newcomer to this field, our goal this year was to investigate the con- tribution of different topic fields in book recommendation. We only considered title, mediated query, group, narrative fields from each topic. We did not consider topic-creator’s catalogue information. Neither we consulted anonymous user pro- files. We submitted five runs (run-ids: ISMD-341, ISMD-342, ISMD-350, ISMD- 354, ISMD-355) in the Suggestion Task. For all the runs, Language modelling 521 with Dirchlet smoothing was used in Lemur’s Indri search system [3]. Our overall performance was not satisfactory. The run with only mediated query was best among our submissions. Organization of rest of the paper is as follows. We describe our approach in Section 2. Section 3 describes dataset and Section 4 reports results. In Section 5 we analyse our results. Finally, we conclude in Section 5 with directions for future work. 2 Approach This year we took a simple approach similar to standard adhoc retrieval. The document collection provided was stopword-removed and then stemmed using Krovetz stemmer. It was indexed with Lemur Indri search system for all the fields having text within. During retrieval, we tried to see the effect of different components of a topic in turn. We therefore used only title (Run-id ISMD-341), only group(Run ISMD- 342), only title with stopword removed (Run ISMD-350), only mediated query (Run ISMD-354), and only narrative with stopword removed field (Run ISMD- 355) from each topic. On top of standard English stopwords we identified a set of a few more like recommendation, hello, suggestion, reference, recent, hi, thank, etc. which we removed in the run ISMD-355. We also removed punctuation marks manually from all the textual content of these fields and used only free text queries in all the runs. We did not consider any other information like catalogue information and user profile during retrieval. For each topic, we submitted 1000 book suggestions in the form of ISBNs. 3 Data Test collection provided by INEX 2014 SBS orgainzers for Suggestion Task had a document collection and a topicset. The document collection consists of 2.8 million book description with metadata from Amazon and LibraryThing. From Amazon there is formal metadata like booktitle, author, publisher, publication year, library classification codes, Amazon categories and similar product infor- mation, as well as user-generated content in the form of user ratings and reviews. From LibraryThing, there are user tags and user-provided metadata on awards, book characters and locations and blurbs. There are additional records from the British Library and the Library of Congress. The entire collection was 7.1 GB in size. [2] The topic-set contains 681 topics each describing a user’s request for sugges- tion of books. Each topic has a set of fields like title, mediated query, group, nar- rative and user’s personal catalogue at the time of topic creation. The catalogue contains a list of book-entries with information like LibraryThing id of the book, its entry-date, rating and tags. 522 The organizers also supplied 94,000 anonymised user profiles from Library- Thing. 4 Results The scores obtained by our five runs are given in Table 1. The official evalua- tion measure by INEX’14 is nDCG@10 [4]. The performance of our runs are in decreasing order. Our best performance is by ISMD-354 where we use only me- diated query field. We also show the best score in the task demonstrated by run- id USTB-run6.SimQuery1000.rerank all.L2R RandomForest(*), for the sake of comparison. Table 1. Results - The official evaluation Measure by INEX 2014 RUN ID Rank M RR nDCG@10 M AP R@1000 ISMD-354 22 0.123 0.067 0.049 0.285 ISMD-341 24 0.106 0.056 0.042 0.236 ISMD-350 27 0.090 0.048 0.036 0.211 ISMD-355 29 0.089 0.038 0.026 0.124 ISMD-342 32 0.018 0.010 0.007 0.081 best* 1 0.464 0.303 0.232 0.390 5 Analysis Although our performance is not up to the mark, there are few take-home lessons. As individual fields, mediated query is the most effective, followed by title and narrative. Removing stopwords from the title is actually detrimental (ISMD-341 and ISMD-350). We did not consider any combination of these fields. It would be interesting to see the performance of different combinations of these fields. 6 Conclusion This year we participated in the Suggestion Task of Social Book Search as ini- tial venture. We tried to see the individual effect of different topic-fields on book recommendation. We considered only a handful of fields like mediated query, ti- tle, narrative etc from the topics. While there can be no denial of the fact that our overall performance is dismal, initial results are suggestive as to what should be done next. We need to consult other fields like book catalogue of the topic creators, ratings of the books in the catalogue during retrieval. We also need to take into account profiles of other users. It is also imperative to see the perfor- mance of combination of different fields in the topics as well as other fields in user catalogues and user profiles. We shall be exploring some of these tasks in the coming days. 523 References 1. Marijn Koolen, Gabriella Kazai, Jaap Kamps, Michael Preminger, Antoine Doucet and Monica Landoni, Overview of the INEX 2012 Social Book Search Track. INEX’12 Workshop Pre-proceedings, Shlomo Geva, Jaap Kamps, Ralf Schenkel (editors), September 17-20, 2012, Rome , Italy. 2. INEX, Initiative for the Evaluation of XML Retrieval. https://inex.mmci.uni- saarland.de/data/documentcollection.jsp 3. INDRI: Language modeling meets inference networks, Available at http://www.lemurproject.org/indri/ 4. Jarvelin, K., Kekalainen, J.: Cumulated Gain-based Evaluation of IR Techniques. ACM Transactions on Information Systems 20(4) (2002) 422-446. 5. INEX, Initiative for the Evaluation of XML Retrieval. https://inex.mmci.uni- saarland.de/ 524