Assignment 3A Natural Language Processing

Assignment 3A
Natural Language Processing ( CS563 )

Instructions:
1. Markings will be based on the correctness and soundness of the outputs.
2. Marks will be deducted in case of plagiarism.
3. Proper indentation and appropriate comments (if necessary) are mandatory.
4. You should zip all the required files and name the zip file as:
<roll_no>_assignment_<#>.zip, eg. 1501cs11_assignment_01.zip.
5. Upload your assignment ( the zip file ) in the following link:
https://www.dropbox.com/request/WMtFjtWFir417V1U2iFc

----------------------------------------------------------------------------------------------------
Sentiment Analysis:
Sentiment analysis is an important task in natural language processing and has a
wide range of real-world applications. The typical sentiment analysis focuses on
predicting the positive or negative polarity of the given sentence(s). This task
works in the setting that the given text has only one aspect and polarity. A more
general and complicated task would be to predict the aspects mentioned in a
sentence and the sentiments associated with each one of them.
As per the above statement, sentiment analysis can be performed in two ways:
A. Sentence-level Sentiment Analysis
B. Aspect-based Sentiment Analysis
(A) Sentence-level Sentiment Analysis
● Problem Statement: For a given message (tweet) classify the sentiment
of the sentence as positive, negative, or neutral.
○ Input: Given Sentence
○ Output: Sentiment of the sentence (positive, negative, or neutral).
● Embedding vectors
○ Use Fasttext embedding for experiment and download the feature
vector from crawl-300d-2M.vec.zip.
○ For more information and usage please go through this link
(https://fasttext.cc/docs/en/english-vectors.html)
● Methods: LSTM
● Dataset: Download the dataset for sentiment classification from here
https://drive.google.com/drive/folders/1_IH0MMD7chR7Sxog0ly7kiUgZoe
5otd_?usp=sharing
○ There will be three files twitter-train.tsv, twitter-dev.tsv, and
twitter-test.tsv in the dataset.
○ Train your model on twitter-train.tsv and validate your model on
twitter-dev.tsv and test your model on twitter-test.tsv
● Evaluation Metrics: Evaluate your model based on the following
metrics:
○ Accuracy
○ F-score
○ Precision and recall
○ Confusion matrix