On Arabic search: The effectiveness of monolingual and bidirectional information retrieval.

摘要

The foremost goal of this research is to develop algorithms for Arabic monolingual and cross-language Information Retrieval (IR) systems to improve the retrieval effectiveness. The inflectional structure of a word has shown a great impact on the performance of the IR systems, namely, the retrieval precision. We present two stemming algorithms for Arabic IR systems. We empirically investigate the effectiveness of the surface-based retrieval. This approach deteriorates the retrieval precision due to the fact that Arabic is a highly inflected language. Accordingly, we propose the root-based retrieval. We notice a significant improvement over the surface-based approach. Many variant word senses are based on an identical root; thus, the root-based algorithm creates invalid conflation classes that result in an ambiguous query. To resolve ambiguity, we propose a light-stemming algorithm for Arabic texts. We show that the light stemming algorithm significantly outperforms the root-based algorithm. We investigate the effectiveness of using automatic relevance feedback technique in Arabic IR systems. We found that automatic relevance feedback achieves superior retrieval effectiveness in an Arabic IR system.; In Cross-Language Information Retrieval (CLIR), queries in one language retrieve relevant documents in other languages. We investigate the Machine Translation (MT) and the Machine-Readable Dictionaries (MRDs) for bidirectional Arabic-English CLIR. The translation ambiguity associated with these resources is the key problem. We present three methods of query translation using a bilingual dictionary for Arabic-English CLIR. First, we present the Every-Match (EM) method. This method yields ambiguous translations. Thus, we present the First-Match (FM) method that considers the first match in the dictionary as the candidate term. Finally, we present a novel translation model called the Two-Phase (TP). We also empirically evaluate the effectiveness of the Arabic-English MT approach using short, medium, and long queries. English-Arabic CUR is evaluated via an MRD and an English-Arabic MT system. The post-translation expansion technique is used to de-emphasize the extraneous terms introduced by MRD and MT for English-Arabic CLIR.