Nada Naji

Information retrieval strategies for digitized handwritten medieval documents

This paper describes and evaluates different IR models and search strategies for digitized manuscripts. Written during the thirteenth century, these manuscripts were digitized using an imperfect recognition system with a word error rate of around 6%. Having access to the internal representation during the recognition stage, we were able to produce four automatic transcriptions, each introducing some form of spelling correction as an attempt to improve the retrieval effectiveness. We evaluated the retrieval effectiveness for each of these versions using three text representations combined with five IR models, three stemming strategies and two query formulations. We employed a manually-transcribed error-free version to define the ground-truth. Based on our experiments, we conclude that taking account of the single best recognition word or all possible top-k recognition alternatives does not provide the best performance. Selecting all possible words each having a log-likelihood close to the best alternative yields the best text surrogate. Within this representation, different retrieval strategies tend to produce similar performance levels.

Back to our roots for retrieving very short passages

This article tackles the task of retrieving very short documents via even shorter queries. The problem on hand may relate to the retrieval of tweets, image and table captions, short text messages (SMS) and sponsored retrieval among others. In such cases, document and/or query expansion using thesauri and other external resources (e.g., Wikipedia) usually available on the World Wide Web (WWW) are proven to be effective approaches. However, the focus of this paper is on documents that are written in lesser known languages for which the WWW is of limited use. Our experiments are based on two main corpora extracted from historical manuscripts written in Latin and Middle High German. We found that retrieving very short documents whose lengths are quite similar via short queries given that no external enrichment resources are available, the classical tf-idf model performs as satisfactorily as the more complex models do, if not better sometimes.