Темир йўл транспорти: долзарб масалалар ва инновациялар 6-jild 2-son (2025) · 86-93-betlar
DATA CLEANING AND PREPROCESSING METHODS FOR AUTOMATIC DETECTION OF TEXTUAL INFORMATION IN SOCIAL NETWORKS: DATA CLEANING AND PREPROCESSING METHODS FOR AUTOMATIC DETECTION OF TEXTUAL INFORMATION IN SOCIAL NETWORKS
Юлдошев, Ю., Отахонова, Б., Обидова, К.
Annotatsiya
Today, social networks, such as Twitter, Facebook, Instagram and other platforms, have become one of the main means of exchanging information between people. Every day, millions of users interact with various text posts, comments, statuses, and thoughts, from which various information, trends, and conclusions can be drawn. At the same time, texts posted on social networks are often written in informal language, with abbreviations, emojis, vague expressions, and sometimes grammatical errors. This makes it difficult to automatically analyze text data. To effectively analyze text data, it is necessary to clean and preprocess it. The process of cleaning data involves removing unnecessary elements from the text, such as special characters, emoji, stop words, and spelling errors. Preprocessing involves formatting the text for analysis, converting words to their root form, and converting them to lowercase. These processes are important for obtaining accurate results in analyzing texts in social networks.
классификация текста, социальные сети, короткий текст, неформальный язык, обработка естественного языка (NLP), машинное обучение, глубокое обучение, токен, стемминг, стоп-слово.text classification, social networks, short text, informal language, natural language processing (NLP), machine learning, deep learning, token, stemming, stop word.
Metadata manbasi: jurnal OAI-PMH arxivi · Sindex toʻliq matnni saqlamaydi, manbaga havola beradi.