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.
I just did my tweets sentiment analysis and I can't share it with anyone. But it's cool knowing that I am not a negative person because the average number is around above 0 tho yet not reaching 1. So here I am sharing with y'all! 😂
He Said linear regression can be used to measure score, based on score sentiment can be classified based on a threshold, e.g. lower score than threshold means negative and vice versa
Martin Sir is a very nice teacher. Thank you for explaining us in a simple manner.
thanks, you are one of the best guys who can explain complex concepts in an easy and smooth way, i really appreciate your efforts
Thank you, IBM.
This really breaks it down into digestible bits.
This was a real information packed thought thumper. Nice job glad I seen the short and your link to it.
Other Companies: thinking about profit
Meanwhile Martin sir and IBM: Lets teach almost each week
These videos are extremely informative and educational--thank you IBM!
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.
I agree. The output of Linear Regression is on continuous spectrum & the output of Logistic regression is discrete (e.g. 0 or 1)
Great video! Thanks for sharing.
very well explained
Nice explanations! Thanks a lot
@IBM Technology please confirm whether it was meant to be logistic regression instead of linear regression
I would sell a Kidney, maybe both, to take courses from Martin on anything computer science, especially AI, related.
Lost in the world of funds - a poetic reflection on the journey to reclaim them.
Thank you so much 😊
Excellent 👌🏻 Explanation
I just did my tweets sentiment analysis and I can't share it with anyone. But it's cool knowing that I am not a negative person because the average number is around above 0 tho yet not reaching 1. So here I am sharing with y'all! 😂
how did he not mention transformers? Please make a video on deploying BERT
One other synonym to lexicon often used by NLP is ontology.
That was perfect
How ontology is used
Linear regression is not a classification method!
He Said linear regression can be used to measure score, based on score sentiment can be classified based on a threshold, e.g. lower score than threshold means negative and vice versa
I think logistic regression is classification method, not linear regression. Because linear reg is predicted continuous values
awesome
Are you writing backwards?
Not really, the camera is behind the board thing and maybe the video is mirrored
Sir 🫡🫡🫡
They can read minds. You guys have been illegally reading my mind via A.I. mind reading technology for over a year now.