|
import numpy as np |
|
from gensim.models import Word2Vec |
|
import gradio as gr |
|
from sklearn.decomposition import PCA |
|
import plotly.graph_objects as go |
|
import nltk |
|
from nltk.tokenize import word_tokenize |
|
from nltk.tag import pos_tag |
|
|
|
nltk.download('punkt') |
|
nltk.download('averaged_perceptron_tagger') |
|
|
|
|
|
def train_word2vec(sentences): |
|
model = Word2Vec(sentences, vector_size=100, window=5, min_count=1) |
|
return model |
|
|
|
def preprocess_text(file_path): |
|
with open(file_path, 'r', encoding='utf-8') as file: |
|
text = file.read() |
|
|
|
|
|
tokens = word_tokenize(text) |
|
tagged = pos_tag(tokens) |
|
|
|
|
|
nouns = [word.lower() for word, pos in tagged if pos.startswith('NN')] |
|
|
|
|
|
unique_nouns = sorted(set(nouns)) |
|
|
|
|
|
sentences = [[noun] for noun in unique_nouns] |
|
|
|
return sentences, unique_nouns |
|
|
|
def apply_pca(word_vectors): |
|
pca = PCA(n_components=3) |
|
return pca.fit_transform(word_vectors) |
|
|
|
def process_text(file_path, target_word): |
|
|
|
sentences, unique_words = preprocess_text(file_path) |
|
|
|
|
|
model = train_word2vec(sentences) |
|
|
|
|
|
word_vectors = np.array([model.wv[word] for word in unique_words]) |
|
|
|
|
|
word_vectors_3d = apply_pca(word_vectors) |
|
|
|
|
|
colors = ['rgba(128, 128, 128, 0.3)' if word != target_word else 'rgba(255, 0, 0, 1)' for word in unique_words] |
|
|
|
|
|
if target_word in model.wv: |
|
similar_words = model.wv.most_similar(target_word, topn=10) |
|
similar_word_indices = [unique_words.index(word) for word, _ in similar_words] |
|
for idx in similar_word_indices: |
|
colors[idx] = 'rgba(0, 255, 0, 1)' |
|
|
|
|
|
fig = go.Figure(data=[go.Scatter3d( |
|
x=word_vectors_3d[:, 0], |
|
y=word_vectors_3d[:, 1], |
|
z=word_vectors_3d[:, 2], |
|
mode='markers+text', |
|
text=unique_words, |
|
textposition="top center", |
|
marker=dict( |
|
size=8, |
|
color=colors, |
|
) |
|
)]) |
|
|
|
fig.update_layout( |
|
title="Word Embeddings 3D Visualization", |
|
scene=dict( |
|
xaxis_title="PCA 1", |
|
yaxis_title="PCA 2", |
|
zaxis_title="PCA 3" |
|
), |
|
width=800, |
|
height=800 |
|
) |
|
|
|
|
|
similar_words_text = "" |
|
if target_word in model.wv: |
|
similar_words_text = "๊ฐ์ฅ ๊ฐ๊น์ด ๋จ์ด 10๊ฐ:\n" + "\n".join([f"{word}: {score:.4f}" for word, score in similar_words]) |
|
|
|
return fig, similar_words_text |
|
|
|
|
|
with gr.Blocks() as iface: |
|
gr.Markdown("# Word Embedding 3D ์๊ฐํ") |
|
gr.Markdown("ํ
์คํธ ํ์ผ(.txt)์ ์
๋ก๋ํ๊ณ ๊ฐ์กฐํ ๋จ์ด๋ฅผ ์
๋ ฅํ์ธ์. Word2Vec๊ณผ PCA๋ฅผ ์ฌ์ฉํ์ฌ ๋จ์ด ์๋ฒ ๋ฉ์ 3D๋ก ์๊ฐํํฉ๋๋ค. ์
๋ ฅํ ๋จ์ด๋ ๋นจ๊ฐ์์ผ๋ก, ๊ฐ์ฅ ์ ์ฌํ 10๊ฐ ๋จ์ด๋ ์ด๋ก์์ผ๋ก ๊ฐ์กฐ๋ฉ๋๋ค. ์ ์ฌํ ๋จ์ด ๋ชฉ๋ก์ ๊ทธ๋ํ ์๋์ ํ์๋ฉ๋๋ค.") |
|
|
|
with gr.Row(): |
|
file_input = gr.File(label="ํ
์คํธ ํ์ผ ์
๋ก๋ (.txt)", file_types=[".txt"]) |
|
word_input = gr.Textbox(label="๊ฐ์กฐํ ๋จ์ด ์
๋ ฅ") |
|
|
|
submit_btn = gr.Button("์ ์ถ") |
|
|
|
plot_output = gr.Plot(label="Word Embedding 3D ์๊ฐํ") |
|
similar_words_output = gr.Textbox(label="์ ์ฌํ ๋จ์ด") |
|
|
|
submit_btn.click( |
|
fn=process_text, |
|
inputs=[file_input, word_input], |
|
outputs=[plot_output, similar_words_output] |
|
) |
|
|
|
if __name__ == "__main__": |
|
iface.launch() |