AI-Based System for Arabic Search Engine

Huge amount of Arabic information and documents are widely spread on the Internet. The accuracy of searching for Arabic documents, as well as other documents, depends on the terms used by the user in the input query to retrieve relevant results and to reduce irrelevant ones. Most results returned are relevant to the query terms not to its meaning. For example, searching for “ ﺗ ﺼ ﻠ ﺐ ا ﻟ ﺸ ﺮ ا ﯾ ﯿ ﻦ – tasallob Al-sharayeen” which means arteriosclerosis will not give results related to “ ا ﻟ ﺠ ﻠ ﻄ ﺔ – Al-jalta” which means clot and “ أ ﻧ ﻮ ا ع ا ﻟ ﺠ ﻠ ﻄ ﺎ ت - anwa’a Al-jaltat” which means types of clots and so on. To solve such problem, this paper proposes an AI system for Arabic search that allowing users to talk to it directly (Ask) questions and Seek clarifications. It will also be able to analyze dynamically hundreds of potential search results / web pages, by starting an Informed – Friendly dialogue and present Innovative answers for the user. Using an interactive AI Arabic search engine with its smart interaction with the users will help finding the best probable results using the least amount of user targeted inquiries.