Mining the treasure of palm leaf manuscripts through information retrieval techniques

摘要

Purpose - This paper aims to popularize information retrieval from palm leaf manuscripts among computer scientists to make available the guidance of the age-old heritage in shaping the future. Design/methodology/approach - With computer technology penetrating every aspect of life, information retrieval algorithms can be exploited to help build a system which can dig into the ocean of knowledge from these manuscripts. Findings - The knowledge in them covers all aspects of life. Be it religious beliefs, literature, science, mathematics, or any other. However, due to discontinuation of practice of copying their content on fresh leaves, they now possess a fragile life which needs to be preserved at the earliest. The modem means of digitization can help in their preservation. Research limitations - The Government of India and other organizations are doing commendable job of preserving and safeguarding country's heritage and age-old knowledge system through the movement of digitization. In the years to come, the agonizing problem of manuscripts degradation will be eradicated completely. However, next when it will come to mining the knowledge treasure out of these manuscripts, we would be confronted with another helpless situation. Practical implications - The digitization process would capture the manuscripts from present physical palm leaf to digital image form by clicking high-quality pictures. All the text in a palm leaf will be available in the form of images, but on these images, a simple search for any word would not be possible. Originality/value - Working towards mining the treasure of knowledge from the palm leaf manuscripts, hordes of challenges have been outlined. Over and above the problem of preventing decay to palm leaf manuscripts is the challenge of deciphering text, image analysis, information retrieval and search. Search is further associated with issues of meaningful and useful extraction through semantic analysis. This paper advocates the dire need for systematic research to be undertaken in this field opening up avenues for past knowledge to guide future prospects in several domains.