IM426 – BUSINESS CASE 6: SOCIAL SENTIMENTAL ANALYSIS 社群情感分析 Original case source & reference: Rainer, Kelly, Prince, Brad and Watson, Hugh, Management Information Systems: Moving Business Forward, John Willey & Sons, Inc.: New Jersey, 3rd Edition, 2015 Prepared by: Celeste Ng Date: March–June 2016
Sentiment analysis 情感分析 Sentiment analysis Refers to the use of natural language processing, text analysis, machine learning, and statistics To identify and extract subjective information in source materials. The expressed opinion can be classified into positive, negative, or neutral.
University of Southern California’s Annenberg Innovation Lab (1) The University of Southern California’s (USC; Annenberg Innovation Lab, researchers are using sentiment analysis To analyze, in real time, the sentiment of conversations on a range of topics that thrive on social media. Researchers hope to help businesses, nonprofit organizations, and government agencies gain new insights from millions of online conversations. The research found that the ability to understand public sentiment in real time was an excellent predictor of how a movie would open (how much money would a movie make in its opening days) and what types of advertising were effective.
Source:
Source: project project
Source: project project
Source:
Translation (1) Sentiment analysis Natural language processing Machine learning Subjective information Neutral Advanced sentiment analysis Emotional states Proliferation Reviews Ratings Reputations Topics that thrive Gain new insights Online conversations Sentiment analysis algorithms Linguistic nuances Sentiment Subject area Politics Entertainment Jargon Sarcasm Sarcastic
Questions 1. Which social sentiment analysis would be beneficial for your university? 2. Potential disadvantage that organizations might experience when using social sentiment analysis are (?) 3. What impacts could social sentiment analysis have on television reality shows?
Translation (2) Enthusiasm Discern Fine-tune A word in quotes New motion picture releases Open days Films Twilight: breaking down Initial excitement The series was ending Conversation Advertising campaign Most talked-about Custom message Show’s programming and advertising Social implications election Public policies Political debate Focus group Dashboard Developing nations Malaria epidemic Civil conflict Proactively Television reality shows
Direct quote, Source: D%A6%E4%B9%A0 D%A6%E4%B9%A0 “ 機器學習是近 20 多年興起的一門多領域交叉學科，涉及機率論、 統計學、逼近論、凸分析、計算複雜性理論等多門學科。機器學 習理論主要是設計和分析一些讓計算機可以自動「學習」的算法。 機器學習算法是一類從數據中自動分析獲得規律，並利用規律對 未知數據進行預測的算法。因為學習算法中涉及了大量的統計學 理論，機器學習與推斷統計學聯繫尤為密切，也被稱為統計學習 理論。算法設計方面，機器學習理論關注可以實現的，行之有效 的學習算法。很多推論問題屬於無程序可循難度，所以部分的機 器學習研究是開發容易處理的近似算法。交叉學科機率論 統計學逼近論凸分析計算複雜性理論計算機學習算法數據規律推斷統計學推論無程序可循難度 機器學習已廣泛應用於數據挖掘、計算機視覺、自然語言處理、 生物特徵識別、搜尋引擎、醫學診斷、檢測信用卡欺詐、證券市 場分析、 DNA 序列測序、語音和手寫識別、戰略遊戲和機器人等 領域。 ”數據挖掘計算機視覺自然語言處理 生物特徵識別搜尋引擎醫學診斷信用卡欺詐證券市 場 DNA 序列語音手寫戰略遊戲機器人