• Place : IPB University (Bogor Agricultural University)
• Course : Dissertation Research

• Background
• Requirement
• Inspiration
• Features

Forest and land fire is an event of forest and/or land burning, either naturally or by human actions, resulting in environmental damage that causes ecological, economic, socio-cultural, and political losses. In 2015, the total area of forest and land fires in Indonesia reached 2,611,411.44 hectares; the area affected by forest fires in 2015 was the worst land and forest fires affected-area from 2013 to 2022. Forest fires cause smog and haze pollution, as well as increasing carbon emissions. Other adverse impacts include forest product degradation and deforestation, the loss of forest products and flood control functions, and the loss of biodiversity. It is necessary to take appropriate field measures of environmental conditions to prevent and manage forest and land fires and to minimize their adverse impacts.

• Python 3.11.2
• Visual Studio Code 1.77.0
• Streamlit 1.19.0
• Jupyter Notebook / Google Colaboratory
• Pandas
• Pillow
• WordCloud
• Matplotlib
• Datetime
• Plotly
• PyPDF2
• NLTK
• Textblob
• Python-Docx
• Openpyxl
• Googletrans 4.0.0-rc1
• XlsxWriter 3.0.8
• Vader Sentiment

• Build Feature Selection Technique (FST) using conventional SSA, PSO, and ALO for analysis of forest and land fires at Indonesia.
• Build Salp Swarm Algorithm-Transfer Function as Feature Selection Technique (FST) for Sentiment Analysis of forest and land fires in Indonesia.
• Build web application to monitor sentiment analysis and Natural Language Processing (NLP) for document uploader and text sentiment input features which support all languages to analyze sentiment of contents related from forest and land fires in Indonesia using Valence Aware Dictionary and Sentiment Reasoner algorithm (new features).

• Post ID : ID number of content post in Twitter
• Conversation ID : ID number of content text conversation tweets
• User ID : ID number of account user that post content tweets
• Created At : Period of content tweet post in Twitter
• Account Name : Account Name of post content tweet in Twitter
• Tweet : Tweet content that have been posted by accoount name
• Mentions : Total of mentions from user that tag the content tweet
• Photos : Total of photos from user that share in Twitter
• Replies Count : Total of replies from user based on content tweets
• Retweets Count : Total of mentions from user that tag the content tweets
• Likes Count : Total of likes from user that like the content tweets
• Hashtags Count : Total of hashtag that used from user to post the content tweets
• Link : Content tweets location in Twitter that have been posted by user
• Sentiment : Content tweet sentiment from user that post from positive, neutral, or negative

• Dinar Ajeng Kristiyanti, B.C.S., M.Com.
• Prof. Dr. Imas Sukaesih Sitanggang, B.Sc., M.Com.
• Dr. Eng. Annisa, B.C.S., M.Com.
• Prof. Dr. Ir. Sri Nurdiati, M.Sc.
• Willibrordus Bayu Nova Pramudya, B.C.S.

The authors would like to thank IPB University for providing support in this research.