Algorithms are important and essential part of computational science research work which is regularly published in scholarly articles. To solve the scaling problem of handling more documents, we examine an intelligent system designed with the goal of automatically identifying algorithms. In this paper, we propose a method to develop an algorithm search engine. The proposed system analyzes a document to discover any algorithm that may be there in the document. If any algorithm is found in the document, the document text is further analyzed to extract additional information about the algorithm. Machine learning and rule based approaches are used to discover algorithm representation. All discovered algorithms and their associated metadata are indexed and offered for searching throughout a text query interface.