Amid the campaign noise, The Times of India employed a large language model (LLM) with a fixed prompt to analyze speeches by three key leaders from April 14–20 in the Tamil Nadu 2026 election, processing a corpus of over 50,000 words.
Methodology
The model broke the text into tokens (words) and detected repeated phrases. A separate prompt mapped the tone of the leaders' speeches by tagging segments into five categories: political, administrative, emotional, populist, and ideological, covering both content and delivery.
Topic Classification
Next, the model classified content into predefined issue topics, including welfare/promises, governance, corruption, price rise/taxes/economics, law and order, political attacks, state rights, and ideology. The topics were limited to six for a common baseline to avoid splitting similar ideas across labels and to maintain classification accuracy.
Speech-o-meter Scoring
Finally, the prompt asked the model to build a "speech-o-meter" to score each leader's speeches on effectiveness using word use, repetition, topics, and tone. A custom analytical framework ensured comparability across leaders. The model assigned weights out of 10: issue dominance (3), recall value (2), repetition efficiency (1.5), emotional connect (1.5), lexical clarity (1), and tone balance (1), which were aggregated to produce the final score.
Get real-time updates and result insights on the UP Board Result 2026 and Assam HS Class 12th Result 2026. You can also check: Gold Rate in Chennai, Silver Rate in Chennai, Bank Holidays in Chennai, Public Holidays in Chennai, Chennai AQI, and Weather in Chennai.
About the Author
Ram Sundaram is an Assistant Editor at The Times of India, Chennai, where he covers commute, trial courts, and political affairs.
