Quest Journals Journal of Software Engineering and Simulation Volume 8 ~ Issue 1 (2022) pp: 06-12 ISSN(Online) :2321-3795 ISSN (Print):2321-3809 www.questjournals.org Corresponding Author: Ofut, Ogar Tumenayu 6 | Page Research Paper Nollywood Movie Sequences Summarization Using a Recurrent Neural Model Ofut, Ogar Tumenayu 1 Department of Computer Science Cross River University of Technology Calabar, Nigeria. Francis Sunday Bakpo Department of Computer Science University of Nigeria Nsukka, Nigeria Debora U Ebem Department of Computer Science University of Nigeria Nsukka, Nigeria Abstract Summarization is the act of presenting the most significant information from an entire work like movie, audio, and text. Movie Sequences Summarization consist of meaningful representation of a movie which mostly enhance effective browsing of large movie collections as well as providing fast content accessing and indexing. In this paper, the propose system is targeted at summarizing Nollywood movie sequences by using a modify Recurrent Neural Network model. Long short-term memory (LSTM) and Bi-directional Long short-term memory (Bi-LSTM) were employed for the enhancement of recurrent neural networks model. The proposed model can condense huge movie and deliver an accurate but brief information about a movie in less time. A custom dataset for Nollywood movie was design following the full protocol of TVsum dataset. Keywords: Summarization; Movie; Dataset; Nollywood; Model; Long short-term memory; and Bi-directional Long short-term memory. Received 04 Jan, 2022; Revised 13 Jan, 2022; Accepted 15 Jan, 2022 © The author(s) 2022. Published with open access at www.questjournals.org I. Introduction Automatic Summarization in general is currently being described as most challenging and interesting problems in the field of Natural Language Processing (NLP). The increasing volume of movie data makes analysis task difficult. Browsing tools are very demanding for the users to obtain a quick information about the movie content. The application of movie sequences summarization model to Nollywood movie industry in place of movie previews to predict the genre of the movie is not just an interesting computer vision problem to solve, but also has extended benefits to stakeholders and decision makers in the industry such as: (i) automatically tagging the genre on content-hosting websites like YouTube and others; (ii) to reduce viewing time, (iii) when categorizing movies, summarization make selection process easier, (iv) automatic summarization improves the effectiveness of indexing, (v) automatic summarization algorithms are less biased than human summarizers, (vi) Personalized summaries are useful in question-answering systems as they provide personalized information [1]. Movie sequences summarization task aims at browsing a set of frames or segments from a visual sequence which contains most significant and informative movie scenes across the entire sequence. Not only is summarization useful for efficiently extracting the substance of data, it also serves many other applications such as video indexing [2], video retrieval [3], and anomaly detection [4]. The Nigerian movie industry called Nollywood is the second highest revenue earner in present day Nigeria. The twenty-first century Nigerian movie industry (Nollywood) produces approximately 2,000 movies a year, which arguably places it in third place on the global movie ranking. Thus, stakeholders in the movie