Challenges in Creating Text Summarization Models in Malayalam: A Study

Automatic text summarization (ATS) is a task where information is being condensed for easy understanding. Malayalam text summarization is in formative state due to the bottleneck of methodologies, datasets and evaluation techniques. Deep learning is a better solution to achieve the balance between output accuracy and system complexity. Through this literature, we are probing into the fundamental aspects of building Malayalam text summarization models so as to highlight prevailing mechanisms, challenges, and future scope in the field. A comprehensive understanding of Malayalam grammatical structure and language-specific feature study has been carried out to achieve the goal. Afterward, available preprocessing tools and their deficiencies were discussed followed by prevailing text summarization mechanisms in Malayalam. Meanwhile, different datasets and evaluation strategies were inquired in order to demarcate their incapacities. This survey would give an overall insight for designing an ATS system in Malayalam, and we hope that the fellow researchers in this area would make use of this road map for literature works.