Computing Reviews

Natural language processing projects: build next-generation NLP applications using AI techniques
Kulkarni A., Shivananda A., Kulkarni A., Apress,New York, NY,2022. 336 pp.Type:Book
Date Reviewed: 11/11/22

Natural language processing (NLP) is now quite widely used in many domains, including commerce, social media, and other text analysis tasks. Paired with artificial intelligence (AI), it enables powerful data analytics tools for prediction, automation, and classification.

Authors Akshay Kulkarni, Adarsha Shivananda, and Anoosh Kulkarni are noted data science practitioners in industry, working to provide commercial applications of AI-powered NLP.

The book’s first chapter opens with a cursory tour of the many concepts and vocabulary of machine learning, data analytics, NLP, and modeling, serving mostly as a reminder to already informed readers rather than instruction and explanation for those new to the field. That is, much is assumed about the target audience’s expertise.

The remaining chapters detail ten specific NLP/AI use cases, covering, for example, sentiment analysis, searching, recommendation, text feature extraction, categorization, and prediction. The authors present the problem statement, approach used, and methodology for each of the projects using Python and its many relevant libraries, including pandas, NLTK, sklearn, Plotly, Matplotlib, Seaborn, NumPy, re, Keras, Gensim, and numerous others. There is a significant amount of minimally commented Python source code that assumes reader fluency in the use of the language and of these libraries. More detailed Jupyter notebooks for each project’s code are available on the book’s GitHub website.

One example project creates a consumer sentiment classifier, adding an emotion detector, listing solution steps along with data preparation, required libraries and functions, and displaying graphical output. Another creates a complex resume screening and ranking tool that includes word clouds for job descriptions and candidates. However, many diagrams and images in the book lack explanatory captions and are poorly scaled and hard to read.

For experienced NLP analysts, the book does provide an interesting collection of example problems and solutions, but such readers will need to exert some effort to extract useful knowledge and practice from the content. Definitely not for beginners.

Reviewer:  Harry J. Foxwell Review #: CR147512

Reproduction in whole or in part without permission is prohibited.   Copyright 2022™
Terms of Use
| Privacy Policy