The average number of daily tweets (average #), together with the mean and standard deviation (s.d.) of tweetwise sentiment scores calculated from the RoBERTa method for different cities in different
1)The data includes raw data (data.xlsx), the stop word dictionary (stop_words.txt), the custom dictionary (my_dict.txt), as well as other reference dictionaries used in natural language processing (d
Sentiment analysis is a pivotal domain in Natural Language Processing (NLP), particularly for understanding opinions expressed in sequential and textual data with the usage of machine learning. It inv
A preprocessed self built sentiment analysis dataset was created by capturing the comments of blogger 'Walking Rice' on Weibo. 0-3 represents no depression, mild depression, moderate depression, and s
