Scribd
Upload
498 views2 pages

Problem Statement - Stock Market News Sentiment Analysis and Summarization - Introduction To Natural Language Processing - Great Learning

The document outlines the requirements and evaluation criteria for a project on Stock Market News Sentiment Analysis and Summarization, with a submission deadline of January 11, 2024. It details various components such as exploratory data analysis, data preprocessing, sentiment analysis using different embedding techniques, content summarization, and actionable insights. The project is graded out of 60 marks, and feedback is provided on the submitted work.

Uploaded by

sanjaycj99
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
498 views2 pages

Problem Statement - Stock Market News Sentiment Analysis and Summarization - Introduction To Natural Language Processing - Great Learning

The document outlines the requirements and evaluation criteria for a project on Stock Market News Sentiment Analysis and Summarization, with a submission deadline of January 11, 2024. It details various components such as exploratory data analysis, data preprocessing, sentiment analysis using different embedding techniques, content summarization, and actionable insights. The project is graded out of 60 marks, and feedback is provided on the submitted work.

Uploaded by

sanjaycj99
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 2

Go Back to Introduction to Natural Language Processing

Course Content

Problem Statement - Stock Market News Sentiment Analysis and Summarization


Submission type : File Upload
Due Date : Jan 11, 1:00 AM PST
Total Marks : 60
Available from : Dec 11, 2024, 10:30 AM
'Your Marks' : 12/60

Description

Scoring guide (Rubric) - Stock Market News Sentiment Analysis and Summarization - Project Rubric Evaluated

Criteria Ratings Points

-- Good work on defining a detailed problem


statement. -- Good work on adding the
observations on shape of data, data types of
various attributes, statistical summary. -- Good
Exploratory Data Analysis work on adding the Univariate and Bivariate
analysis (variable distributions, interactions
- Problem definition - Univariate analysis - Bivariate between variables) to understand the
analysis - Use appropriate visualizations to identify the relationships in data. -- Good work adding the 8
patterns and insights - Key meaningful observations insights based on EDA. - Good work on adding
on individual variables and the relationship between comments on range of attributes, outliers of
variables various attributes. - Good work on commenting
on the distribution of the variables and potential
fixes for better modelling. -- Good work on
adding the key insights shared from univariate
and bivariate analysis.

Data Preprocessing -- Good work on splitting the target variable and


- Split the target variable and predictors - Split the predictors. -- Good work on splitting the data 2
into train, validation, and test sets.
data into train, validation, and test sets

Word2Vec -- Good work on defining the model.


-- Good work on encoding the data. -- Average
Word Embeddings the word vectors to get sentence encodings.
GloVE -- Good work on defining the model. --
- Using Word2Vec - Using GloVe - Using Sentence 10
Good work on encoding the data. - Average the
Transformer word vectors to get sentence encodings.
Sentence Transformer -- Good work on defining
the model. -- Good work on encoding the data.

Sentiment Analysis -- Word2Vec embeddings 1) Good work on 16


building a base model. 2) Good work on
- Comment on which metric to use and why - Build a performing hyperparameter tuning. 3) Good
ML model with each of the following embedding work on commenting the model performance.
techniques - Word2Vec - GloVe - Sentence GloVe embeddings 1) Good work on building a
Transformer - Perform hyperparameter tuning for the base model. 2) Good work on performing
hyperparameter tuning. 3) Good work on
ML model for each of the following embedding commenting the model performance. Sentence
techniques - Word2Vec - GloVe - Sentence Transformer embeddings 1) Good work on
Transformer - Comment on model performance across building a base model. 2) Good work on
Criteria Ratings Points

different metrics - Choose the best model from the performing hyperparameter tuning. 3) Good
ones built with proper reasoning - Check the work on commenting the model performance. --
performance of the final model on the test set Note: Compared the performances of all the models
and choose the final model with proper
The ML model to use is at the discretion of the learner. rationale. -- Good work on checking the final
model's performance on the test set.
Improvements: -- Add comment on which metric
to use and why.

Content Summarization
- Group the data at a week-level - Load the large -- Good work on grouping the data at a week-
language model from Hugging Face - Create a level. -- Summarization 1) Load the model from
function to define the model parameters and generate Hugging Face. 2) Create a function to define the
a response - Define the instruction for the task (the model parameters and generate a response. 3)
task is to identify the top three positive and negative Define the instruction for the task. 4) Apply the 12
response generation function to get an output
events from the week that are likely to impact the from the model. 5) Create a DataFrame
stock price) - Apply the response generation function containing the necessary fields from the
to get an output from the model - Create a DataFrame model's output in a structured manner.
containing the necessary fields from the model's
output in a structured manner

Actionable Insights and Recommendations


-- Good work on adding the Actionable Insights.
- Share your observations and insights from the - Good work on adding the Recommendations 4/4
analysis conducted - Provide recommendations for the for the problem.
business

Presentation/Notebook - Overall Quality


- Structure and flow - Crispness - Visual appeal - All - The notebook looks well structured with a
key insights and recommendations covered? OR - logical flow - The notebook looks clean and well 8/8
commented code.
Structure and flow - Well commented code - All key
insights and recommendations covered?

Points 12/60

Submitted Assignment

BruceKyle_NLP_Project_Full_Code.fixed.html

Submitted at Mon, Jan 06

Comments:

Good job overall! Please check the comments on individual sections. Keep up the good work. Best wishes!
Dr VRK, Jan 19, 11:53 AM

+ Add comments

© 2013 - 2025 Great Learning Education Services Pvt. Ltd. All rights reserved
Privacy · Terms

You might also like