SUMMARY
Public perception on climate change is a paramount component that affects the implementation of adaptation and mitigation measures. Taking into account the public perceptions on the issue may assist decision-makers in producing appropriate strategies to ameliorate the impacts of climate change. A corpus-driven sentiment analysis approach was done to classify the polarity of Malaysian public perceptions, identify the sentiment lexicon, and analyse the public sentiments. A part of a specialised corpus namely the Malaysian Diachronic Climate Change Corpus (MyDCCC) was developed from The Sun Daily and was used as the data for this study. The methodology involved the employment of Azure Machine Learning software to conduct sentiment analysis to explore the polarity of public sentiments, corpus analysis approach to identify the sentiment lexicon and discourse analysis to analyse public sentiments based on the identified sentiment lexicon. The results revealed that the majority of public sentiments appeared to be negative, depicting sentiment words such as long, critical, and serious. Positive sentiment words also prevailed such as better, best and hope. The discourse analysis revealed that the public is reasonably insightful of climate change although their sentiments appeared to be negative. However, the negative stance was largely influenced by the public's indignation with how decision-makers handle the climate change issue. Ironically, the negative sentiments may be an indication for the decision-makers to improve their approach in addressing climate change. This study has contributed significantly to research on public perceptions of climate change in the Malaysian context.