Experiments with Different Indexing Techniques for Text Retrieval tasks\n on Gujarati Language using Bag of Words Approach

This paper presents results of various experiments carried out to improve\ntext retrieval of gujarati text documents. Text retrieval involves searching\nand ranking of text documents for a given set of query terms. We have tested\nvarious retrieval models that uses bag-of-words approach. Bag-of-words approach\nis a traditional approach that is being used till date where the text document\nis represented as collection of words. Measures like frequency count, inverse\ndocument frequency etc. are used to signify and rank relevant documents for\nuser queries. Different ranking models have been used to quantify ranking\nperformance using the metric of mean average precision. Gujarati is a\nmorphologically rich language, we have compared techniques like stop word\nremoval, stemming and frequent case generation against baseline to measure the\nimprovements in information retrieval tasks. Most of the techniques are\nlanguage dependent and requires development of language specific tools. We used\nplain unprocessed word index as the baseline, we have seen significant\nimprovements in comparison of MAP values after applying different indexing\ntechniques when compared to the baseline.\n