Abstract: In this span of erudition of information, Universities can lead competitive advantage of searching of resources only by trained data analysis. This paper highlights context free data cleaning for improved tag cloud by correcting values of user defined “Tags”, using different string similarity metrics, where “Tags” are assigned by users which related to referenced resource. Authors propose a procedure to scrutinize suitability of value to correct other values of Tags. Several string similarity metrics were used, to find distance of two different strings and generate results. Experimental results show how the approach can meritoriously clean the data without reference data.

Keywords: Context free data cleaning; Information Retrieval; String similarity metrics; Tag Cloud.