Digital Report Grading Using NLP Feature Selection

Autor: R. Shiva Shankar, D. Ravibabu
Rok vydání: 2018
Předmět:
Zdroj: Soft Computing in Data Analytics ISBN: 9789811305139
Popis: The trend of online examination has been so common in today’s world for the ones testing their English proficiency. As the number of persons increasing to take the tests such as GRE, TOEFL, GMAT that involves the essay writing. Due to the lack of efficient, grading of these essays by the graders led to the automatic evaluation of essays and textual summaries. It is a hectic for the graders to provide feedback with stable interface, mindset, and time bounds. Features like bag of words, sentence, and word count along with their average length, structure, and organization of an essay are used for achieving maximum accuracy in grading. The sequential forward feature selection algorithm is used to compare the accuracy and select the best subset in this way. An efficient subset is formed from a single empty set because of this algorithm and its operations made it easy to implement and perform well on small data sets.
Databáze: OpenAIRE