Skip to content

tthsep17/social-media-listening-NLP

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 

Repository files navigation

social-media-listening-NLP

Motivation & Business Problem:

The number of internet users has exploded tremendously in the last decade as a result of technological advancements. The amount of data generated by people's engagement on the internet has increased dramatically. Amidst the noise is a wealth of knowledge regarding customers’ opinions and feedback that we can tap into. Apple's core values have always been prioritizing people over products. We're eager to hear from users through social media listening, a new way to analyze and collect customers' thoughts via natural language processing. Particularly, we are interested in answering these questions:

  1. How do the customers feel generally towards the products?
  2. What attributes drive customers' interest in purchasing the product?
  3. Could we predict whether a person adopts/uptakes the product based on the post?

To conduct this, we will mainly be looking at the social media data in 2017, when our major competitor, Samsung, launched their Samsung Galaxy S8 flagship product. Later in the same year, we also launched iPhone 8 and iPhone X. The time frame provides us with a fair foundation for comparison. Regarding the sources, we have collected data from the following sources: Instagram, Tumblr, Facebook, Twitter, blogs, forums, comments, consumer reviews, and other social media.

The whole project pipeline will be divided into 8 steps:

  • Step 0: Load the package
  • Step 1: Load the File
  • Step 2: Data Preprocessing
    • 2.1 Remove Irrelevant Rows
    • 2.2 Summary Statistics
    • 2.3 Exploratory Data Analysis
    • 2.4 Feature Engineering: Category and Before/After Launch
    • 2.5 Handle Null Values
    • 2.6 Preprocessing Text Data
  • Step 3: Demographic Analysis
    • 3.1 Overview of Demographics
    • 3.2 Compare US vs. Non-US
  • Step 4: Sentiment Analysis
    • 4.1 Overall Sentiment
    • 4.2 Sentiment Before VS After Launch
  • Step 5: Attribute Analysis
    • 5.1 Topic Modeling
    • 5.2 POS Analysis
    • 5.3 Hashtag Frequency
    • 5.4 Sentiment Analysis for Product Attributes
  • Step 6: Predictive Analysis
    • 6.1 Feature Engineering
    • 6.2 Prediction of Product Uptake
    • 6.3 Model Fitting
  • Step 7: Comparison between Twitter and non-Twitter Dataset
    • 7.1 Exploratory Data Analysis of Twitter and Non-Twitter Dataset
    • 7.2 Word Frequency for Twitter and Non-Twitter Dataset
    • 7.3 Hashtag for Twitter and Non-Twitter Dataset
    • 7.4 Sentiment Analysis for Product Attributes for Twitter and Non-Twitter Dataset
  • Step 8: Recommendation
    • 8.0 Preparation: Attribute Enhancement
    • 8.1 Introduction: Informative Marketing
    • 8.2 Growth: Persuasive Marketing: Persuasive Marketing
    • 8.3 Maturity: Product Consolidation
    • 8.4 Decline: Pivoting

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published