Description

This LP is divided into four parts.

• The first aims to trace a short history of Sentiment Analysis from the origin, in the 1950s, to nowadays, where it is widely used to mine personal information from content on the Internet. In this regard, as Bo Pang and Lillian Lee argued: “the year 2001 or so seems to mark the beginning of widespread awareness of the research problems and opportunities that sentiment analysis and opinion mining raise […] and subsequently there have been […] hundreds of papers published on the subject” (p.9)
• The second part will focus on Sentiment Analysis, also known as Opinion Mining, which is “an active research area in natural language processing”, as stated by Duyu Tang, Bing Qin and Ting Liu in their paper Deep learning for sentiment analysis. It aims to build a process of interpretation of users’ emotional attitude on the web within text data, and in particular “at identifying, extracting and organizing sentiments from user generated texts in social networks, blogs or product reviews” (p.1) making use of NLP, text analysis and other machine learning systems. One of the main tasks of this kind of analysis is to identify the polarity of a given text, to understand whether the opinions expressed within it are positive, negative or neutral, which reveals to be incredibly useful when companies use it to survey the popularity of their products.
• The third part shows how Lords of Data operate over the internet to interpret our emotions and will. In the digital world, Lords of Data use information and data to identify customer sentiment toward products, brands or services in online conversations and feedback. By applying sentiment analysis, Lords of Data can monitor social media, brand feedback, market researches and use all these data to personalize advertisments and improve digital markets or businesses. Scott Galloway's talk about how giant digital markets and social media platforms such as Amazon, Apple, Google or Facebook manipulate and profile our emotions can be an example of relation between today's digital world and sentiment analysis.
• The last part focus on twitter data. Through two video lectures that show how the data are collected and the different ways of the way of analyzing different models to classify Tweets posted into positive, negative and neutral sentiments.

Type of learning path

sequence

Path nodes

1. • Sentiment Analysis brief history and use today
2. • History of Sentiment Analysis from 2001 to nowadays
3. • Introduction to Sentiment Analysis
4. • Deep learning for sentiment analysis: successful approaches and future challenges
5. • NLP - Linear Models for Text Sentiment Analysis
6. • NLP - Linear Models for Text Sentiment Analysis - Transcription
7. • A Quick Guide To Sentiment Analysis
8. • How sentiment analysis lets you find out what your customers like (and dislike)
9. • Lords of Data
10. • EU ‘data sovereignty’: opportunities and challenges
11. • Why we should have agency over our digital identity data & how to get it
12. • How Amazon, Apple, Facebook and Google manipulate our emotions | Scott Galloway
13. • How Amazon, Apple, Facebook and Google manipulate our emotions | Scott Galloway - Transcription
14. • Sentiment Analysis On Social Media
15. • Sentiment Analysis Of Twitter Data

Submitted by Narmin Safarova
21/03/2020
in the project 4. What sentiment analysis is today? How Lords of Data profile our emotions

last updated by ANTONELLA ROSSINI -- 03/06/2020