This paper describes the work that we did at Indian School of Mines towards Social Book Search Track for INEX 2014. We submitted ve runs in its Suggestion Task. We investigated individual e ect of title, group, mediated query, and narrative elds of the topics in our runs. For all the runs we used language modelling technique with Dirichlet smoothing. The run using only mediated query eld was our best. Overall, our performance is not satisfactory. However, as new entrant to the eld, our scores are encouraging enough to work for better results in future.
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
enable users, Social Book Seach Track at INEX [
As a newcomer to this eld, our goal this year was to investigate the contribution of di erent topic elds in book recommendation. We only considered title, mediated query, group, narrative elds from each topic. We did not consider topic-creator's catalogue information. Neither we consulted anonymous user proles.
We submitted ve runs (run-ids: ISMD-341, ISMD-342, ISMD-350,
ISMD354, ISMD-355) in the Suggestion Task. For all the runs, Language modelling
with Dirchlet smoothing was used in Lemur's Indri search system [
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
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 elds having text within.
During retrieval, we tried to see the e ect of di erent components of a topic in turn. We therefore used only title (Run-id ISMD-341), only group(Run ISMD342), only title with stopword removed (Run ISMD-350), only mediated query (Run ISMD-354), and only narrative with stopword removed eld (Run ISMD355) from each topic.
On top of standard English stopwords we identi ed 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 elds and used only free text queries in all the runs.
We did not consider any other information like catalogue information and user pro le during retrieval.
For each topic, we submitted 1000 book suggestions in the form of ISBNs. 3
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 classi cation codes, Amazon categories and similar product
information, 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. [
The topic-set contains 681 topics each describing a user's request for suggestion of books. Each topic has a set of elds like title, mediated query, group, narrative 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.
The organizers also supplied 94,000 anonymised user pro les from LibraryThing. 4
The scores obtained by our ve runs are given in Table 1. The o cial
evaluation measure by INEX'14 is nDCG@10 [
This year we participated in the Suggestion Task of Social Book Search as initial venture. We tried to see the individual e ect of di erent topic- elds on book recommendation. We considered only a handful of elds like mediated query, title, 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 elds like book catalogue of the topic creators, ratings of the books in the catalogue during retrieval. We also need to take into account pro les of other users. It is also imperative to see the performance of combination of di erent elds in the topics as well as other elds in user catalogues and user pro les. We shall be exploring some of these tasks in the coming days.