Building a Multi-lingual Field Association Terms Dictionary

摘要

Field association terms (FATs) dictionary concept was studied and tested for English. All algorithms were developed considering the English rules, power link algorithm presented a new rules to improve the quality of field association terms (FATs) dictionary in English. This method is independent of the Language then it can be applied to any language. Arabic language has many differences from the English language so it needs special techniques for preprocessing before applying the power link algorithm. In this paper depending on available FATs dictionary in English a multilingual FATs dictionary in English and Arabic is presented. The results showed that the precession and recall for the Arabic dictionary were less than the English one. Recall and precision for the Arabic FATs dictionary were 87.7% and 83.6%. In our simulation the power link algorithm increased average precision for the Arabic FATs were by 18.6%. Also the complete dynamic system to build a multi lingual FATs dictionary was built and tested.