Reputation: 303
When considering Social Networks like Twitter and Facebook, what can be the mechanism / technique used to do predictions for "age" and "gender"? Can it be done through Machine Learning or Sentiment Analysis or both?
Upvotes: 1
Views: 145
Reputation: 768
There has been research working on the automatic detection of user latent variables, including age and gender, on social media data. These studies have taken into account a variety of features and evaluated their effectiveness. For instance, the content of one's social media post can say a lot about their age and gender. For example, studies have shown that if someone uses the word 'buddy', the user is more likely to be a young male.
So, the answer to your question is 'Yes'. You can use machine learning techniques to detect age and gender on social media. However, choosing an effective set of features depends on the context you want to study and the platform you want to focus on, and it requires some experimentation. Use of sentiment as a feature might be useful in one context and might not be of help in another.
I refer you to the following articles that have studied this topic before:
Upvotes: 2
Reputation: 303
Machine learning is a part of artificial intelligence where your algorithms learn on (usually big) data. It subdivides into classification, regression, clustering and other disciplines.
Natural language processing can use machine learning but it can be also engineered by hand.
Sentiment analysis is a part of NLP. It usually uses machine learning (classification).
When considering Social Networks like Twitter and Facebook to do predictions for "age" and "gender", sentiment analysis can be used and it's a ML mechanism used in NLP.
Upvotes: 1