When we retrieve information from a website or web-based system we receive both relevant and unwanted material. Reading unwanted material wastes time and reduces productivity. A well-designed text analyzer can help minimize this problem. This research presents a text analyzer that reduces the amount of unwanted material retrieved. It filters unwanted material by using the knowledge that it had gained previously or acquired by grouping data elements. Results show that the accuracy of information retrieved is proportional to the efficiency of work done and decisions made. Its performance is compared and bench-marked with other text analyzers. Initial investigations based on several sample runs show that this text analyzer is more efficient than many others.

Text analysis, text analyzer, information retrieval, web-based systems, short documents, text mining