Brain Sentiment Indicator for Sectors

The objective is to provide a sentiment on sectors starting from the calculation of stock sentiment from financial news. The sentiment scoring technology is based on a combination of various natural language processing techniques. The sentiment score assigned to each stock is a value ranging from -1 (most negative) to +1 (most positive) and it is updated with a daily frequency. The sentiment score corresponds to the average of sentiment for each piece of news and it is available on two time scales; 7 days and 30 days. The sentiment of each sector is then calculated as an average of the sentiment based on financial news of all stocks belonging to the sector for various time intervals. The sentiment calculation process is the following: 1. Financial news are collected every few minutes from various financial media 2. Brain engine assigns a specific category to each piece of news (e.g. “patent win” or “contract lose”) using semantic rules. Each category has a predefined value of sentiment. 3. If the categorization fails a bag of words approach is used based on dictionaries customized for Financial news. The approach includes a strategy for negation handling. 4. Repetition of similar news is kept into account in the sentiment aggregation. 5. The sentiment is aggregated at stock level and then at sector level

搜集汇总

背景与挑战

背景概述

该数据集旨在基于金融新闻计算股票情感得分，从而生成行业情感指标。情感得分通过自然语言处理技术（如语义规则分类或词袋方法）对新闻进行分析，范围在-1到+1之间，每日更新并提供7天和30天的平均结果。最终，情感从股票层面聚合到行业层面，并考虑了新闻重复因素。