Information Retrieval Technology: 6th Asia Information Retrieval Societies Conference, AIRS 2010, Taipei, Taiwan, December 2010. Proceedings

Citation

du Preez, M. (2011), "Information Retrieval Technology: 6th Asia Information Retrieval Societies Conference, AIRS 2010, Taipei, Taiwan, December 2010. Proceedings", Online Information Review, Vol. 35 No. 6, pp. 976-977. https://doi.org/10.1108/14684521111193265

The papers presented in these proceedings of the sixth Asia Information Retrieval Conference (AIRS) represent the work by academics and practitioners from across the world. Twenty‐six quality papers on a diverse range of topics related to information retrieval technology are divided into eight sections. These cover information retrieval models, machine learning for information retrieval, user studies, natural language processing for information retrieval, multimedia, and web‐related topics. The volume also includes 30 papers that were presented at a poster session. These papers cover a diverse range of topics that include the use of ontologies and folksonomies, relevance ranking paradigms, cross‐lingual information retrieval, quantitative and qualitative analysis of web search processes, etc.

Cyber crime is a great concern to both law enforcement agencies and those involved in web development projects. One example of a cyber‐crime activity that concerns researchers is the availability and distribution of extremist videos on the web. In their paper Sureka and colleagues describe a semi‐automated system to assist agencies dealing with cyber‐crime activities related to the promotion of hate and radicalism on the Internet. Their description explains how they successfully mined manually validated hate videos on YouTube and, through social network analysis techniques, were able to identify some virtually hidden communities.

With a personal interest in user studies and multimedia, the papers in these two sessions interested me. The user studies and evaluation session consisted of three papers. The first paper discussed an effective time‐ratio system which can be used to evaluate search engines' performance. The authors deployed a real user study and showed that effective time‐ratio reflects users' satisfaction better than the existing metrics based on document relevance.

Information system users tend to end their information searches without clicking on any results. This tendency was the focus of the study by Yao and colleagues. Their findings are very useful for user behaviour reliability studies. In the concluding paper for this session, Ravana and Moffat explored some design options that should be considered when planning a comparative evaluation of information retrieval systems.

The multimedia session concludes the volume. The first of two papers proposes a music retrieval algorithm based on users' emotion tags. The authors report on experimental results which promise to be better than the traditional cosine similarity and the co‐tags similarity methods currently being used. The second and last paper proposes an approach to re‐ranking images retrieved from existing image search engines. The authors conducted several experiments on datasets sampled from INRIA Holiday datasets, Photo.net, DPChallenge, Google Image, and Flickr.

This volume is packed with high quality papers and will be most useful for computer and information technology specialists interested in various aspects of information retrieval.