Integrating Social Features and Query Type Recognition in the Suggestion Track of CLEF 2015 Social Book Search Lab Shih-Hung Wu1*, Yi-Hsiang Hsieh1, Liang-Pu Chen2, Tsun Ku2 1 Chaoyang University of Technology, Taiwan, R.O.C { shwu(*Contact author), s10027006,}@cyut.edu.tw 2Institute for Information Industry, Taiwan, R.O.C eit@iii.org.tw, cujing@gmail Abstract. The Social Book Search (SBS) Lab is part of CLEF 2015 lab series. This is the third time that the CYUT CSIE team attends the SBS track. Based on a full-text search engine, we build a social feature re-ranking system and in- troduce more knowledge on understanding the queries. We defined a set of rules to filtering out unnecessary books from the recommendation list. The official run results show that the system performance is improved from our previous system. Keywords: Query type recognition, social features, social book search 1 Introduction The paper reports our system in the suggestion track of CLEF 2015 Social Book Sug- gestion (SBS) [10]. This is the third time that we attend the SBS track since 2013 INEX [7]. Based on our social feature re-ranking system [1], we improve our system by involving some knowledge on understanding the queries. We believe that the result of traditional information retrieval technology is not enough for the users who need more personal recommendation in the SBS task. Rec- ommendation from other users are more appealing; it might contain more personal feelings and cover more subtle reasons that traditional information retrieval system cannot cover. Our system integrates the social feature into the traditional information retrieval technology to give better recommendation on books. In this task, user- generated metadata is used as the social feature. According to our observation on the topics in the previous INEX SBS Track, we found that queries can be separated into different types. Simply treating the keywords in the topic as search terms will not get good results. Some queries require higher level of knowledge to deal with. System needs to understand the information need behind the keyword, for example, the knowledge on the types of literature. We analy- sis the topics and find several types in them. Due to the time limitation, we only im- plement a module to recognize one special type of topics and a filtering module to modify the recommendation result. The structure of this paper is as follows. Section 2 is the data set description, sec- tion 3 shows our architecture and the details of our method, section 4 is the experi- ment results, and final section gives conclusions and future works. 2 Dataset 2.1 Collection The document collection in this task is provided by the CLEF 2015 Social Book Sug- gestion track. The documents are the XML format metadata of about 2.8 million books and the data size is 25.9GB. These documents are collected from Amazon.com and LibraryThing [2]. The XML tags used in the data set is listed in Table 1. Table 1.All the XML tag [2] tag name book similarproducts title imagecategory dimensions Tags edition name reviews Isbn dewey role editorialreviews Ean creator blurber images Binding review dedication creators Label rating epigraph blurbers Listprice authorid firstwordsitem dedications manufacturer totalvotes lastwordsitem epigraphs numberofpages helpfulvotes quotation firstwords publisher date seriesitem lastwords Height summary award quotations Width editorialreview browseNode series Length content character awards Weight source place browseNodes readinglevel image subject characters releasedate imageCategories similarproduct places publicationdate url tag subjects Studio data 2.2 Test Topic Topics provided by CLEF 2015 Social Book Suggestion track are collected from Li- braryThing. A topic describes the information needed for a user. Figure 1 and Figure 2 give partial view of an example, the XML tags used are：, , <medi- ated_query>, <group>, <narrative>, <catalog>, <book>, <LT_id>, <entry_date>, and <rating>. Where title means the title of a post on LibraryThing forum and narrative is the content of the post. While mediated_query is added as an interpretation of the query. Group means the user group in the forum of the user who post this query. <topics> <topic id="1196"> <title>The Best Peace Corps Novel books about work for Peace Corps Returned Peace Corps Volunteer Readers I'm looking for people's concept of what is the best novel for the Peace Corps Volunteer - pre, during, o r post service. This could be a novel that typifies life in th e country of service. It could be a novel that typifies the wo rk volunteers do. It could be a novel that makes for the perfe ct reading while in service. Anything will do, just give rea sons. It might lead other PCVs/RPCVs to interesting reading. Let's try novels, and then head into non-fiction later... I'l l start: I could not have survived my 2 years of service if I had not read Chingiz Aitmitov's The Day Lasts More than A Hun dred Years and Bulgakov's The Master and Margarita . They re ally made most of my concerns about my own sanity living in th e crumbling remnants of Soviet Central Asia vanish into vapor, as I was able to learn that not only was surreality the norm for this part of the world but also my own preconceptions abou t the concrete, rational world that I thought I knew might be questionable. Figure 1. A topic example in CLEF 2015 Social Book Suggestion track 120241 yes positive 10151 yes positive 42437 2006-07 10.0 1270696 2006-07 10.0 ... Figure 2. A topic example in CLEF 2015 Social Book Suggestion track (Continued) 3 CYUT CSIE System Methodology 3.1 System Architecture Figure 3 shows our system architecture. The pre-processing modules include stop words filtering, and stemming, both modules are provided by Lucene [6]. After the preprocessing, our system builds index for retrieval. The results of content-based retrieval will be re-ranked as the final results according to the social features. 3.2 Indexing and Query The index and search engine in use is the Lucene system, which is an open source full text search engine provided by Apache software foundation. Lucene is written in JAVA and can be called easily by JAVA program to build various applications. Table 1 shows all the tags of the book metadata. According to Bogers and Larsen [3], there are 19 tags more useful in the social book search. They are , , <publisher>, <editorial>, <creator>, <series>, <award>, <character>, <place>, <blurber>, <epigraph>, <firstwords>, <lastwords>, <quotation>, <dewey>, <sub- ject>, <browseNode>, <review>, and <tag>. Our system also focuses on the same 19 tags. In the pre-processing step, the content in the <dewey> tag is restored to strings ac- cording to the 2003 list of Dewey category descriptions [9] to make string matching easier. For example: <dewey>004</dewey> will be restored to <dewey>Data pro- cessing Computer science</dewey>. The content of <tag> is also expanded according to the count number to emphasize its importance. For example: <tag count="3">fantasy</tag> will be expanded as <tag>fantasy fantasy fantasy</tag>. In additional to the 19 tags, our system also indexes the content of <review> as inde- pendent indexes files and names it as reviews. Fig.1 and 2 shows all the XML tags of the query topics. According to Koolen et al. [4], an Indri [5] based system using all the contents of <Title>, <Query>, <Group>, and <Narrative> as query terms will give better result. We also use the contents of the four tags as our system input queries. Document Stop words Collection Stemming filtering Content-based Indexing Query Retrieval Type2 Search Engine Filtering Yes No Search Engine Re-Ranking Results Figure 3. System architecture <topic id="76778"> <title>Russian Serfdom Suggestions Russian serfdom History Readers: Clio's (Pleasure?) Palace I'm reading Flashman At The Charge right now and Russian serfdom is a prominent feature. Any one have any good suggestions to learn more abo ut this aspect of Russian history during the Ts ars? I'm looking for a Gulag: A History about serfdom. Thanks! Figure 4. A type2 query example that we defined in 2015 SBS track 3.3 Type2 Query Recognition and Result Filtering According to our observation on the topics in INEX 2012 SBS Track, we find that there are some queries that are different from others, we call them the Type2 queries [11]. Type2 queries are the queries that contain the names of some books that the orig- inal users want to find similar ones. Therefore, the books in the topics should not be part of the recommendation. Since the book names are given explicitly, our system originally will find exactly the same books as the top recommendation. To recognize type2 queries, we define a list of phrases to identify such queries and filter out the books in the queries from the recommendation lists. The phrases are listed in the ap- pendix in the previous paper [11]. Figure 4 gives an example of Type2 queries taken from INEX 2013 SBS topics, in which contains a key phrase “I’m reading”. We find that there are 174 queries in the INEX 2013 SBS track that can be classified as Type2 queries. Therefore, we add a module in our system to identify the Type2 queries and filtering out the books mentioned in the topics. 3.4 Re-ranking The Re-ranking part is similar to that in our previous work [1]. We integrate the user- generated metadata into the traditional content-based search result by re-ranking the results. The social features are provided by the amazon users, and our system use them to give more weight on certain books. Three numbers are available:  User rating: users might evaluate a book from 1 to 5, the higher the better.  Helpful vote: other users might endorse one comment by voting it as helpful.  Total vote: the total number of helpful or not. We designed 3 different ways to use these social features in re-ranking. 1) User rating method Increase the weight of content-based retrieval result by adding the summation of user rating. As shown in formula (1): Scorere−ranked (i) = α ∗ Scoreorg (i) + (1 − α) ∗ Scoreuser rating (i) (1) 2) Average User rating method Increase the weight of content-based retrieval result by adding the average of us- er rating. As shown in formula (2): Scorere−ranked (i) = Scoreorg (i) + Scoreaverage user rating (i) (2) 3) Weights User rating method Increase the weight of content-based retrieval result by adding the book which gets more helpful votes. As shown in formula (3) and (4): helpfulvote ScoreWeights User Rating = User rating ∗ (3) totalvote Scorere−ranked (i) = α ∗ Scoreorg (i) + (1 − α) ∗ ScoreWeights User Rating (i) (4) 3.5 Find the Best α Value by Experiment Since there is no theoretical reference on how to set the α value, in our official runs, the value is selected via a series experiments that we conduct on the 2013 dataset. Table 2 shows the results, we find that the system gets the best result when α is 0.95. Table 2. Experimental Result for different α on 2013 data set Α P@10 MAP 0.50 0.0221 0.0193 0.60 0.0221 0.0193 0.70 0.0224 0.0195 0.80 0.0226 0.0196 0.90 0.0237 0.0204 0.95 0.0245 0.0220 4 Experimental results In the official evaluation, we sent four runs. We use four fields in the topics as query terms, and we filter out some book candidates for all the type2 queries. The configura- tion of each run is as follows.  Run 1, the CSIE - 0.95AverageType2QTGN, re-ranking with Average User Rating.  Run 2, the CSIE - Type2QTGN: without re-ranking.  Run 3, the CSIE - 0.95RatingType2QTGN, re-ranking with User Rating.  Run 4, CSIE - 0.95WRType2QTGN, Re-ranking with Weights User Rating. According to Table 2, the parameterα is 0.95 for best result in the runs with re- ranking. Table 3 shows the official evaluation results of our four runs. Among them the CSIE - 0.95AverageType2QTGN run gives the best NDCG@10 [8] result, while the CSIE - Type2QTGN run gives similar result on NDCG@10 but give better result on MAP and R@1000. The other two runs give poorer results might due to technical errors. Comparing to the 2013 INEX SBS results in Table 5, our system performance improved significantly. However, comparing to the result of INEX SBS 2014 in Table 4, our system performance decreased. Table 3. Official evaluation results in 2015 SBS Run nDCG@10 MRR MAP R@1000 Profiles CSIE - 0.95AverageType2QTGN 0.082 0.194 0.050 0.319 no CSIE - Type2QTGN 0.080 0.191 0.052 0.325 no CSIE - 0.95RatingType2QTGN 0.032 0.113 0.019 0.214 no CSIE - 0.95WRType2QTGN 0.023 0.072 0.015 0.216 no Table 4. Official evaluation results in 2014 INEX SBS Run nDCG@10 MRR MAP R@1000 CYUT - Type2QTGN 0.119 0.246 0.086 0.340 CYUT - 0.119 0.243 0.085 0.332 0.95AverageType2QTGN CYUT - 0.95RatingType2QTGN 0.034 0.101 0.021 0.200 CYUT - 0.95WRType2QTGN 0.028 0.084 0.018 0.213 Table 5. Official evaluation results in 2013 INEX SBS Run nDCG@10 P@10 MRR MAP Run1.query.content-base 0.0265 0.0147 0.0418 0.0153 Run2.query.Rating 0.0376 0.0284 0.0792 0.0178 Run3.query.RA 0.0170 0.0087 0.0352 0.0107 Run4.query.RW 0.0392 0.0287 0.0796 0.0201 Run5.query.reviwes.content- 0.0254 0.0153 0.0359 0.0137 base Run6.query.reviews.RW 0.0378 0.0284 0.0772 0.0165 5 Conclusions and Future work This paper reports our system and result in CLEF 2015 Social Book Suggestion track. We sent four runs and the formal run results are list in Table 3. In the four runs, the CSIE - 0.95AverageType2QTGN run gives best nDCG@10, which is searching with content-based search engine, applying a set of filtering rules based on a list of key phrase and re-ranking with Average User Rating. In the future, we will implement more modules with literature knowledge on the writers, genre of books, geometric categories of the publishers, and temporal categories of the authors that can deal with the special cases in the topics. From this year, user profiles are available, which can be used to give better recom- mendation. A system might use the user profiles to expand the queries or to suggest more books that the user read before for other similar users. Outside resources might also be used to expand the queries. For example, a system might check Wikipedia to find more authors of the books in the same genre, and make better recommendation. Books that won some awards might also be a good list for recommendation. Acknowledgement “This study is conducted under the "Online and Offline integrated Smart Commerce Platform(2/4)" of the Institute for Information Industry which is subsidized by the Ministry of Economy Affairs of the Republic of China . References 1. Wei-Lun Xiao, Shih-Hung Wu, Liang-Pu Chen, Hung-Sheng Chiu, and Ren-Dar Yang, “Social Feature Re-ranking in INEX 2013 Social Book Search Track”, CLEF 2013 Evalua- tion Labs and Workshop Online Working Notes, 23 - 26 September, Valencia, Spain. 2. 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,P.77-P.96,2012. 3. Toine Bogers and Birger Larsen, “RSLIS at INEX 2012: Social Book Search Track”, INEX'12 Workshop Pre-proceedings,P.97-P.108,2012. 4. Marijn Koolen, Hugo Huurdeman and Jaap Kamps, “Comparing Topic Representations for Social Book Search”, CLEF 2013 Evaluation Labs and Workshop Online Working Notes, 23 - 26 September, Valencia – Spain. 5. T. Strohman, D. Metzler, H. Turtle, and W. B. Croft, “Indri: a language-model based search engine for complex queries”, In Proceedings of the International Conference on Intelligent Analysis, 2005. 6. Lucene, https://lucene.apache.org 7. Marijn Koolen, Gabriella Kazai, Michael Preminger, and Antoine Doucet, “Overview of the INEX 2013 Social Book Search Track”, CLEF 2013 Evaluation Labs and Workshop Online Working Notes, 23 - 26 September, Valencia – Spain. 8. Järvelin, K., Kekäläinen, “J.: Cumulated Gain-based Evaluation of IR Techniques”, ACM Transactions on Information Systems 20(4) (2002) 422–446. 9. 2003 list of Dewey category descriptions, https://www.library.illininois.edu/ugl/about/dewey.html 10. CLEF 2015 Social Book Search Track, http://social-book- search.humanities.uva.nl/#/suggestion 11. Shih-Hung Wu, Pei-Kai Liao, Hua-Wei Lin, Li-Jen Hsu, Wei-Lun Xiao, Liang-Pu Chen, Tsun Ku, and Gwo-Dong Chen Query Type Recognition and Result Filtering in INEX 2014 Social Book Search Track 12. Marijn Koolen, Toine Bogers, Gabriella Kazai, Jaap Kamps, and MichaelPreminger Over- view of the INEX 2014 Social Book Search Track