Cleaning and Mapping Interface

With the growth in data day by day, it is getting harder to collect data in a clean format. Data collected for different purposes from sources contains more noise than the data required for the purpose. This research focuses on the cleaning of data and making it meaningful to use by building an annotator which reduces manual labor and makes annotation more efficient. I built a Cleaning and Mapping Interface which detects misspelled word and corrects it, even helps in predicting the next words based on its previous knowledge. It can be used by users with no knowledge of any library or functional programming to annotate their data. It is a user-friendly software that is easy to use. The application is written in Python language with a simple user interface implemented using the Tkinter module. The program has the ability to learn new grammar. The algorithm is recursively applied to deal with the same sentences (that are matched before correcting them) in the document, making it less time consuming and efficient. Cleaning and Mapping Interface uses algorithms for recurrent neural networks and Naive Bayes for making predictions. It provides the user with correct spelling and the next word for a respective word, so annotation can be done faster.