INTELLECTUALIZATION OF PROJECT MANAGEMENT WEB SERVICES BASED ON INTEGRATION WITH NATURAL LANGUAGE PROCESSING MODULES

Abstract

The article explores a possibility of project management web services intellectualization and perfection with the use of natural language processing techniques. It is shown, that the support of high reliability, extensibility, flexibility and density of the interface of such systems is a necessary condition for the providing of effective tools for developers of software systems of various profiles. The directions of development of modern means and services of natural language processing are analyzed. These include libraries and toolkits such as Google Cloud Natural Language API, Amazon Comprehend, IBM Watson's, CVLizer, Wit.ai, DialogFlow, TextRazor, InData Labs API, Paralleldots API, NLP фреймворк BitText, Ambiverse API, Datumbox API and Indico.io. The requirements for the project management system are formulated in terms of reliability, extensibility, flexibility, interface density and availability. The article shows the advantages of using natural language processing tools to improve the efficiency and availability of project management web services. Since most of natural language processing public web services provide similar functionality with similar performance, it is reasonable to create a multiservice integral system architecture that would be able to access several such services at concurrently (based on load or quality criteria), while falling back to the locally deployed natural language processing system when it comes to processing secure or confidential data or in case of external services failures.