What is Sentiment Analysis?

How do companies analyze your online texts to gauge sentiment, using techniques from NLP and machine learning? In this video, Martin Keen explains the power of sentiment analysis and explores the nuances, pitfalls, and benefits of sentiment analysis for improving customer experiences and brand reputation.

[aninditachakraborty]

Martin Sir is a very nice teacher. Thank you for explaining us in a simple manner.

[ahmadsaud]

thanks, you are one of the best guys who can explain complex concepts in an easy and smooth way, i really appreciate your efforts

[ShoyomboRaphael]

Thank you, IBM.
This really breaks it down into digestible bits.

[osznnum]

Other Companies: thinking about profit

Meanwhile Martin sir and IBM: Lets teach almost each week

[bikedawg]

These videos are extremely informative and educational--thank you IBM!

[bikedawg]

FROM CHATGPT:
In sentiment analysis using a machine learning approach, logistic regression is the appropriate classification algorithm, not linear regression. Here's why:

*Logistic Regression*

Classification Task:
Sentiment analysis is typically a binary classification task (positive vs. negative sentiment) or a multi-class classification task (e.g., positive, neutral, negative).
Logistic regression is designed for classification problems, making it suitable for sentiment analysis.

Output:
Logistic regression outputs probabilities that can be mapped to discrete classes (e.g., sentiment labels). The sigmoid function is used to convert the linear combination of input features into a probability between 0 and 1.
The model predicts the class label based on the highest probability.


Decision Boundary:
Logistic regression creates a decision boundary to separate different classes based on the feature space, which is essential for classification tasks.


*Linear Regression*

Regression Task:
Linear regression is used for regression problems, where the goal is to predict a continuous outcome variable.
It is not suitable for classification tasks because it predicts a continuous value rather than discrete class labels.


Output:
Linear regression outputs a continuous value, which is not directly interpretable as class labels for sentiment analysis.


Decision Boundary:
Linear regression does not create a decision boundary for classification. Instead, it fits a line (or hyperplane) to predict continuous values.

[Patricia-yPatricia_x]

Lost in the world of funds - a poetic reflection on the journey to reclaim them.

[seanurquhart]

I would sell a Kidney, maybe both, to take courses from Martin on anything computer science, especially AI, related.

[oshkit]

@IBM Technology please confirm whether it was meant to be logistic regression instead of linear regression

[glenilame]

how did he not mention transformers? Please make a video on deploying BERT

[sammyfrancisco]

One other synonym to lexicon often used by NLP is ontology.

[evandrogoulart]

Linear regression is not a classification method!

[ahmedshaikh]

They can read minds. You guys have been illegally reading my mind via A.I. mind reading technology for over a year now.