PROJECT PROPOSAL - Emotion Detection from Text

[Ank1taS]

Emotion Detection from Text

A model will predict the emotion behind a sentence i.e model will take a sentence or group of statements and classify it to a specific emotion class

---

Field	Description
About	Using a dataset available on the Kaggle ML model will predict the emotion/ state of mind of the writer who wrote the text segment.
Github	Ank1taS
Email	sahoosona2209@gmail.com
Label	Project Request

https://github.com/Ank1taS

Define You

• [x] GSSOC Participant
• [x] Contributor

Emotion Detection from Text

Description

This ML model will take a sentence or group of statements and classify it to a specific emotion class. I wanted to identify the emotion behind the words written.

It will classify into classes like -

• Anger
• Fear
• Happy
• Love
• Sad
• Surprise

Also, I plan to compare the accuracy of different models over the data set.

Scope

This model can integrate with applications or websites which ultimately ease the review system of comments or feedback. Limitation - As the Model predicts emotions based on the words of the sentence hence I can not identify sarcasm.

Timeline

Required 5 - 8 days as I am new to Machine Learning.

Video Links or Support Links

Data set

[vanshsharma5503]

Hey @Ank1taS Can I be part of this project?

[Ank1taS]

Hello! @vanshsharma5503 Yes sure but can you please give me a brief idea of how would you contribute, and to what part? However, Let's wait for the admins to accept the proposal and assign me to it 😄

[vanshsharma5503]

@Ank1taS thanks for giving me opportunity. As I have just started with ml. So I don't have clear idea how I can ideate. If can u help me like how to start it would be really greatful.

[Ank1taS]

@vanshsharma5503 lets wait @adithya-s-k to assign the issue ✌️

[geyani998]

Hey @Ank1taS @vanshsharma5503 @adithya-s-k Could you please assign me this issue, I would follow the following steps to do this:

To classify emotions in sentences,

1) I will Collect labeled data: I would gather a dataset of sentences labeled with emotions like anger, fear, happy, love, sad, and surprise.

2) Prepare the data: I will clean the sentences by removing punctuation and converting everything to lowercase, and then split the sentences into individual words to get them ready for modeling.

3) Model Selection: I would select a machine learning model suitable for text classification, like Naive Bayes classifier, Random Forest, ANN, Multilinear perception, and many more. 

4) I will train and evaluate the model: I would split the labeled dataset into training and testing sets. Train the model on the training data, adjusting its internal parameters to predict the correct emotion class. Finally, I will evaluate the model's performance on the testing set by calculating accuracy, precision, recall, and F1 scores. 

Please give a chance

[ghost]

Hello, my name is Ishanya. I wish to contribute to this project, I like the idea behind the project, could kindly assign me the project? Thank You