Hands On Program on Text and Social Sentimental Analytics

With the popularity of Social Media like Facebook, Blogs, Twitter and Interactive Company Web sites for interaction with Customers and Suppliers, a large volume of Digital Data has been collect- ed. However, not all of this data is being put to decision making. The majority of these data are in the form of Text, Video, and Images etc. They are also collectively known as unstructured data, as they do not comply with standard definition of Databases. Analyzing Unstructured data and extracting relevant information like Key Words, Topics and also automatically classifying the document from what is also known as Text Analytics. The Text Analytics market is projected to grow approximately from US$3.2 billion in 2016 to about US$8.8 billion by 2022 with a CAGR of about 17%.

Sentiment Analysis is a type of unstructured data analysis. It is a combination of Natural Language Processing, Statistics & Machine Learning to identify and extract subjective information from text. Some of the examples are Reviews of Products, Stock market Sentiment, Digital Marketing, E-Commerce, Dynamic pricing Employee Satisfaction Surveys, Social & Legal sector data. The information, so extracted, will be combined with other predictive analytic techniques such as Regression, Decision Trees to improve the Quality of Decision Making and consequently add to the bottom line.

This workshop will be fully hands-on, with real life data. Minimum theoretical concepts will be covered as and when required along with the Hands On Sessions. Open source GUI based software KNIME will be used, which is very powerful for Text and Image Processing software. KNIME has been placed as a leader for Data Science and Machine Learning Platforms in Gart- ner’s Magic Quadrant. 2020

Program Objectives

  1. Appreciate the challenges in handling   unstructured data
  2. Integration of different sources of unstructured data from Blogs, Website Print, etc.
  3. How to extract  information extracted from these sources?
  4. Decision Making from the extracted information.
  5. Incorporation of sentiment information with predictive models

Learning Outcomes

  1. Unstructured data integration from different sources
  2. Different data cleaning and preprocessing techniques adopted
  3. Sentiment Extraction
  4. Document Classification
  5. Data visualisation techniques like Word Map, Cloud Map, Tag Cloud etc

Modules of the Program

Hands-on sessions covering the above concepts using KNIME Open Source.Software with real-life data sets in the area of Finance, Marketing, Customer Survey Analysis etc.

1. Introduction to Text Mining:

a. Challenges in Handling Text Data, Video data and Web
b. Language Role
c. Overview of English Language – Parts of Speech (POS)
d. Overview of POS of some of common Taggers used in Text Processing

3. Text Transformation:

a. Feature Extraction using 1 Word/2 Word Pairs
b. Term Frequency/Inverse Document Frequency
c. Word2Vector
d. N-Gram Frequencies

2. Text Pre-processing:

a. Document Creation and MetaData Extraction
b. Named Entity Tagging
c. Location Tagging
d. Parts of Speech Tagging
e. Word Stemming
f. Punctuations Filtering and Stop Word Filtering

4. Introduction to Neural Networks & Decision Trees in context of Text Classification:

a. Term Frequency and Inverse Document Frequency Analysis
b. Document Clustering and Classification
c. Sentiment Extraction
d. Topic Modeling
e. Output: Cloud Map Word Map etc.

Target Audience

  1. Business Analysts,
  2. Data Scientists
  3. Faculty/Research Staff of Educational Institutions.
  4. Corporate MIS professionals

Date :- 11th & 12th March, 2022

Timings:- 5 PM to 8 PM – ( 30 minutes break time after 90 minutes )

Venue:- Online

Charges INR 9999 ( All inclusive per participant )

Total  Participants – 35 Nos Max 

Scroll Up