Tamil Summary Generation for a Cricket Match

Cricket is one of the most followed sports in the Indian subcontinent. There is a wide requirement for natural language descriptions, which summarize a cricket match effectively. The process of generating match summaries from statistical data is a manual process. The objective of this paper is to propose a framework for automatic analysis and summary generation for a cricket match in Tamil, with the scorecard of the match as the input. Data analytics is performed on the statistical match data, to mine all frequently occurring patterns. The paper proposes a parameter called Interestingness, which quantifies the interestingness of the match. The paper also proposes a customization model for the summary. We propose an evaluation parameter called humanness, which quantifies the similarity between the output and a manually written summary. Discussing the results and analyzing the summaries generated for matches based on scorecards, this paper concludes with proposing some extensions for future developments.