Scribd
Upload
10 views4 pages

Gensim: A Python Library For NLP and Word Embeddings.: Import As Import As From Import From Import From Import

The document outlines a program that uses dimensionality reduction techniques like PCA to visualize word embeddings from a selected domain, specifically sports. It includes code to generate semantically similar words for a given input using GloVe embeddings. Additionally, it provides examples of visualizing word relationships and printing vector embeddings for specific words.

Uploaded by

Shamanth M
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
10 views4 pages

Gensim: A Python Library For NLP and Word Embeddings.: Import As Import As From Import From Import From Import

The document outlines a program that uses dimensionality reduction techniques like PCA to visualize word embeddings from a selected domain, specifically sports. It includes code to generate semantically similar words for a given input using GloVe embeddings. Additionally, it provides examples of visualizing word relationships and printing vector embeddings for specific words.

Uploaded by

Shamanth M
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/ 4

PG2: Use dimensionality reduction (e.g., PCA or t-SNE) to visualize word embeddings for PG 1.

Select
10 words from a specific domain (e.g., sports, technology) and visualize their embeddings. Analyze
clusters and relationships.

Generate contextually rich outputs using embeddings. Write a program to generate 5 semantically
similar words for a given input

Soln:

!pip install gensim

#Gensim: A Python library for NLP and word embeddings.


Use dimensionality reduction (e.g., PCA or t-
SNE) to visualize word embeddings for PG1. Select 10 words from a specific domain (e.g., sp
orts, technology) and visualize their embeddings. Analyze clusters and relationships. Genera
te contextually rich outputs using embeddings. Write aprogram to generate 5 semantically si
milar words for a given input.

import numpy as np
import matplotlib.pyplot as plt
from sklearn.decomposition import PCA
from sklearn.manifold import TSNE
from gensim.models import KeyedVectors

# Load pre-trained GloVe embeddings (100d model)


model_100d =
KeyedVectors.load_word2vec_format("/content/glove.6B.100d.word2vec.txt"
, binary=False,limit=500000)

# Select 10 words from a specific domain (sports) # Included other


words to show how embeddings are different
words = ['football', 'soccer', 'basketball',
'tennis','engineer','information', 'baseball', 'coach', 'goal',
'player', 'referee', 'team']
word_vectors = np.array([model_100d[word] for word in words])

# Dimensionality reduction using PCA


# Using PCA to reduce to 2D for visualization
pca = PCA(n_components=2)
pca_result = pca.fit_transform(word_vectors)

# Plotting the words in 2D space


plt.figure(figsize=(10, 8))
for i, word in enumerate(words):
plt.scatter(pca_result[i, 0], pca_result[i, 1])
plt.text(pca_result[i, 0] + 0.02, pca_result[i, 1], word,
fontsize=12)
plt.title("PCA Visualization of Sports-related Word Embeddings (100d)")
plt.xlabel("PCA Dimension 1")
plt.ylabel("PCA Dimension 2")
plt.show()

# 5 Semantically Similar Words Generator Function


def get_similar_words(word, model, topn=5):
similar_words = model.similar_by_word(word, topn=topn)
return similar_words

# Example: Get 5 words similar to "football"


similar_words_football = get_similar_words('football', model_100d,
topn=5)
print(f"Words similar to 'football': {similar_words_football}")

Output:
Output: Words similar to 'football': [('soccer', 0.8732221722602844),
('basketball', 0.8555637001991272), ('league', 0.815336287021637),
('rugby', 0.8007532954216003), ('hockey', 0.7833694815635681)]

# Select the words you want to print embeddings for


words_to_print = ['football', 'soccer']

# Print their embeddings


for word in words_to_print:
if word in model_100d:
print(f"Vector embedding for '{word}':\n{model_100d[word]}\n")
else:
print(f"Word '{word}' not found in the embeddings model.")

Output:

Vector embedding for 'football':


[ 0.43865 0.10537 0.45972 -1.0724 -1.2471 0.76351
0.47528 0.083857 -0.9127 -0.27328 -0.018591 -1.184
0.22748 0.16847 -0.52158 0.11339 1.3757 0.11892
-0.37683 0.51149 -0.8833 0.96259 0.18143 -0.407
0.036181 -0.74432 -0.0027401 -0.70068 0.53103 0.45114
-0.72884 1.0631 -0.28008 -0.63848 0.15645 -0.46927
-1.0071 1.033 -1.4354 -0.27485 0.048984 0.13951
0.43072 -0.78791 0.41097 0.58509 1.0155 -0.1839
0.27487 -0.90866 -0.30441 -0.17396 0.020941 0.62813
0.10978 -2.3885 -0.56364 -0.27193 0.98728 0.70608
-0.512 0.52636 -0.78503 -0.68714 0.38121 0.097582
-0.20237 0.43208 -0.30527 0.57925 0.62619 -0.47415
0.33834 -0.28421 -0.097465 0.19597 0.54849 0.59918
-0.41576 0.1021 0.6766 0.0042009 -0.12354 -0.76613
-0.27436 -0.68248 -1.0789 -0.16708 0.81671 0.026999
-0.38707 0.40448 -1.0995 0.64718 -0.12802 -0.26084
-0.96701 0.88078 1.012 -0.022223 ]

Vector embedding for 'soccer':

[ 8.3777e-01 5.1890e-01 6.4015e-01 -6.2606e-01 -9.7474e-01 1.0127e+00

6.2729e-02 4.4316e-01 -8.3299e-01 7.9888e-02 -1.1815e-02 -1.1265e+00

1.2554e-01 -3.4206e-01 -5.1422e-01 3.8526e-01 1.0032e+00 -1.5172e-03

-2.2684e-01 3.5658e-01 -6.2449e-01 8.7271e-01 3.6670e-01 4.6462e-01

-1.0046e-01 -4.4798e-01 -2.1813e-01 -5.6423e-01 5.6665e-01 5.1601e-01

-5.6511e-01 7.1919e-01 -6.5347e-01 -9.5952e-02 5.6028e-01 -4.9956e-01

-7.4757e-01 6.8516e-01 -1.4518e+00 -1.1207e-01 1.0241e-01 3.0537e-02


1.1326e-02 -8.6873e-01 6.3622e-01 4.9539e-01 3.0538e-01 7.7133e-02

7.4048e-02 -7.1163e-01 -1.9159e-01 -3.4168e-01 -4.7185e-01 5.6794e-01

3.7454e-01 -1.9207e+00 -8.6040e-01 5.7058e-01 1.0700e+00 9.2101e-01

-6.4825e-01 5.3516e-01 -1.5556e-01 -9.0021e-01 -1.7459e-01 3.3146e-02

-5.7512e-01 2.9963e-01 -4.0008e-01 -1.0765e-01 4.1384e-01 -7.2178e-01

1.1442e-01 -2.1291e-01 5.4949e-02 1.3213e-01 7.8766e-01 8.9291e-02

-6.6689e-01 3.3998e-01 9.7163e-01 -8.4871e-02 1.7542e-01 -4.6039e-01

-8.5885e-02 -7.5960e-01 -1.5071e+00 2.1545e-01 2.1209e-01 -4.4837e-01

-2.5882e-01 3.3814e-01 -4.7979e-01 2.1059e-01 2.3621e-01 -3.6699e-01

-8.1440e-01 5.4515e-01 9.7946e-01 2.3367e-01]

You might also like