International Journal of Applied Information Systems |
Foundation of Computer Science (FCS), NY, USA |
Volume 12 - Number 30 |
Year of Publication: 2020 |
Authors: Ifeanyi-Reuben Nkechi J., Odikwa Henry, Benson-Emenike Mercy E. |
10.5120/ijais2020451859 |
Ifeanyi-Reuben Nkechi J., Odikwa Henry, Benson-Emenike Mercy E. . The Performance Evaluation of an Igbo Text-Based Intelligent System. International Journal of Applied Information Systems. 12, 30 ( June 2020), 1-5. DOI=10.5120/ijais2020451859
The expansion in Information Technology (IT) has inculcated Igbo, one of the three Nigeria major languages in text-based intelligent systems such as text classification, data/information retrieval and natural language processing. The evaluation of an intelligent system and its processes is very important for the progress of the system. This paper presents a performance evaluation of an Igbo text-based intelligent system. A system performance evaluation is the procedure by which a system’s resources and results are measured to find out if the system is operating at an optimal level. The performance evaluation of the system was done on classification results of an Igbo text represented with Unigram and Bigram Language Models. Object-Oriented design methodology is used for the work and is implemented with the Python programming language with tools from Natural Language Toolkit (NLTK). The system performance is assessed by calculating the precision, recall and F1-measure of the classification result obtained on Unigram, Bigram and Trigram represented text. Result shows classification on an Igbo Bigram represented text has higher level of precision and accuracy.