The rapid diffusion of social media like Facebook and Twitter, and the massive use of different types of forums like Reddit, Quora, etc., is producing an impressive amount of text data every day.
There is one specific activity that many business owners have been contemplating over the last five years, that is identifying the social sentiment of their brand, by analysing the conversations of their users.
In this episode I explain how one can get the best shot at classifying sentences with deep learning and word embedding.
Additional material
Schematic representation of how to learn a word embedding matrix E by training a neural network that, given the previous M words, predicts the next word in a sentence.
Word2Vec example source code
https://gist.github.com/rlangone/ded90673f65e932fd14ae53a26e89eee#file-word2vec_example-py
References
[1] Mikolov, T. et al., "Distributed Representations of Words and Phrases and their Compositionality", Advances in Neural Information Processing Systems 26, pages 3111-3119, 2013.
[2] The Best Embedding Method for Sentiment Classification, https://medium.com/@bramblexu/blog-md-34c5d082a8c5
[3] The state of sentiment analysis: word, sub-word and character embedding
https://amethix.com/state-of-sentiment-analysis-embedding/
Chatting with ChatGPT: Pros and Cons of Advanced Language AI (Ep. 215)
Accelerating Perception Development with Synthetic Data (Ep. 214)
Edge AI applications for military and space [RB] (Ep. 213)
From image to 3D model (Ep. 212)
Machine learning is physics (Ep. 211)
Autonomous cars cannot drive. Here is why. (Ep. 210)
Evolution of data platforms (Ep. 209)
[RB] Is studying AI in academia a waste of time? (Ep. 208)
Private machine learning done right (Ep. 207)
Edge AI for applications in military and space (Ep. 206)
[RB] What are generalist agents and why they can change the AI game (Ep. 205)
LIDAR, cameras and autonomous vehicles (Ep. 204)
Predicting Out Of Memory Kill events with Machine Learning (Ep. 203)
Is studying AI in academia a waste of time? (Ep. 202)
Zero-Cost Proxies: How to find the best neural network without training (Ep. 201)
Online learning is better than batch, right? Wrong! (Ep. 200)
What are generalist agents and why they can change the AI game (Ep. 199)
Streaming data with ease. With Chip Kent from Deephaven Data Labs (Ep. 198)
Learning from data to create personalized experiences with Matt Swalley from Omneky (Ep. 197)
State of Artificial Intelligence 2022 (Ep. 196)
Create your
podcast in
minutes
It is Free
Insight Story: Tech Trends Unpacked
Zero-Shot
Fast Forward by Tomorrow Unlocked: Tech past, tech future
The Unbelivable Truth - Series 1 - 26 including specials and pilot
Well There‘s Your Problem