Implementation Of Movie Database Query Using Natural Language Processing & Python
Understanding natural language is hard. Even in human-to-human interaction there are miscommunications. Learning the SQL database query language and constructing a well-formed SELECT statement to retrieve data can also be challenging. The goal of this work is the connecting of a natural language query in human form relating to movies and actors to a database that persists the corpus and provide the user with information relevant to the question they asked. This negates the need for end-user technical database expertise. It combines the simplicity and flexibility of Python with the easily accessible language processing features of the open source Natural Language Tool Kit (NLTK) library.
By representing a sample of movie database queries in human form with NLTK grammar rules, and matching of rule nouns and verbs with a WordNet corpus of synonyms, the design shows how a larger number of meanings can be derived and transformed into SQL code generation from a compact set of grammar rules, facilitating rapid development and simplifying rule management whilst improving the overall robustness, usability and success of the application.
Master’s Degree Intelligent Systems paper grade 81% | Source code
A Review Of State Of The Art Using Prolog For NLP
Prolog is a declarative general-purpose logic programming language first conceived in the early 1970’s by French computer scientist Alain Colmerauer. Its roots are in first order predicate calculus, and based on facts and rules. It has been widely used in artificial intelligence, especially within the interdisciplinary field of Natural Language Processing (NLP), providing applications with the ability to process human language due to its suitability in assembling human linguistics models for language search, determination and generation.
This report presents a review of the state of the art using Prolog with emphasis on practical application within NLP over the last 5 years.