Hello, My name is Seprian Najib Hilmi

Performed data crawling, preprocessing, and analysis across multiple platforms including Google News, YouTube, TikTok, X (Twitter), Instagram, and Threads.

Utilized a combination of web scraping tools (Selenium & BeautifulSoup) and LLMs such as ChatGPT API and DeepSeek to assist with data preprocessing—automating tasks like text cleaning, summarization, entity extraction, and sentiment analysis.

Built scalable pipelines for structured and unstructured data to support downstream analysis such as topic modeling, trend detection, and social media monitoring.

Conducted comprehensive sentiment analysis of Twitter data to assess public perceptions of property companies, with a focus on business units in Karawang and Bandung. Through in-depth interviews with over 50 consumers, I gained valuable insights to inform future strategies and improve customer engagement.

Utilizing SPSS software, I analyzed over 1,000 data points to extract actionable insights, contributing to the development of effective business strategies. Additionally, I developed secondary research databases from Dapo Kemendikbud, supporting internal property news and decision-making processes.

Collected and processed news data related to Food Estate Merauke and Pesut Conservation initiatives from various local and national news portals (2021–2023). Utilized Python for data crawling, pre-processing, and cleaning to ensure structured, high-quality datasets ready for analysis.

Conducted sentiment analysis on the Food Estate discourse using a pre-trained Indo-BERT model to uncover public opinion trends and emotional tones across time and media sources.

Applied Social Network Analysis (SNA) and Text Network Analysis (TNA) using tools like Nvivo and Gephi to identify key actors, themes, and discourse structures. These findings offered strategic insights for stakeholders involved in both agricultural development and environmental conservation projects.