Biodiversity Knowledge Retrieval Application Using Natural Language Processing Technique

Biodiversity knowledge can be considered as complex information that involves several levels including genes, species, and ecology. As the data is complex, it may exist in form of unstructured and unorganized. The ways researchers collect their data according to their interest needs can worsen this situation because there are no standard ways of doing it. Also, the imperfect collaboration between the biological scientist and the information technology community can result in inconsistent data retrieval. These problems can be overcome by the development of retrieval application as it assists in managing and retrieving biodiversity information. The collection of biodiversity data is stored in the database. By doing this way, it will be easier to be managed as it is kept in the most structured way. For retrieving the information, this application received the user’s query as an input. The query will go through several natural language processing techniques including tokenization of a query, removing the stop words, and lemmatizing the generated tokens to produce better output for the user. The process will be repeated until the user is satisfied with the output he or she is looking for. Several test cases have been done in the testing phase and this application has proven to work accordingly based on the conducted analysis. Recall and precision metrics were also being conducted and the result is quite promising. Further research is recommended for producing a better result by improving the query processing techniques and adding more data to the database.