File size: 4,978 Bytes
0d82e3a 3d8c480 0d82e3a 3d8c480 0d82e3a |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 |
import gradio as gr
import pandas as pd
import numpy as np
import networkx as nx
import matplotlib.pyplot as plt
from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity
import plotly.graph_objects as go
from joblib import load
# Load the data and models
df = pd.read_csv('netflix_titles.csv')
node_embeddings = load('node_embeddings.pkl')
# Create TF-IDF matrix
def create_soup(x):
features = []
if isinstance(x['description'], str):
features.append(x['description'])
if isinstance(x['cast'], str):
features.append(x['cast'])
if isinstance(x['director'], str):
features.append(x['director'])
if isinstance(x['listed_in'], str):
features.append(x['listed_in'])
if isinstance(x['title'], str):
features.append(x['title'])
return ' '.join(features)
df['combined_features'] = df.apply(create_soup, axis=1)
tfidf = TfidfVectorizer(stop_words='english')
tfidf_matrix = tfidf.fit_transform(df['combined_features'])
cosine_sim = cosine_similarity(tfidf_matrix, tfidf_matrix)
def get_hybrid_recommendations(query, n_recommendations=10):
# Convert query to TF-IDF vector
query_vec = tfidf.transform([query])
# Get content-based similarity scores
content_scores = cosine_similarity(query_vec, tfidf_matrix)[0]
# Combine scores with weights
weights = {'content': 0.7, 'node_embeddings': 0.3}
final_scores = content_scores * weights['content']
# Get top recommendations
sim_scores_with_index = list(enumerate(final_scores))
sim_scores_with_index = sorted(sim_scores_with_index, key=lambda x: x[1], reverse=True)
sim_scores_with_index = sim_scores_with_index[:n_recommendations]
# Create recommendations DataFrame
recommendations = []
for i, score in sim_scores_with_index:
recommendations.append({
'title': df['title'].iloc[i],
'type': df['type'].iloc[i],
'similarity_score': score,
'description': df['description'].iloc[i],
'genres': df['listed_in'].iloc[i]
})
return pd.DataFrame(recommendations)
def create_bar_chart(recommendations_df):
fig = go.Figure(data=[
go.Bar(
x=recommendations_df['title'],
y=recommendations_df['similarity_score'],
marker_color=recommendations_df['similarity_score'],
text=recommendations_df['similarity_score'].round(3),
textposition='auto',
)
])
fig.update_layout(
title='Top 10 Recommendations',
xaxis_title='Title',
yaxis_title='Similarity Score',
xaxis_tickangle=-45,
height=500
)
return fig
def create_network_graph(recommendations_df):
G = nx.Graph()
plt.figure(figsize=(12, 8))
# Add nodes
for _, row in recommendations_df.iterrows():
G.add_node(row['title'], type=row['type'], score=row['similarity_score'])
# Add edges
titles = recommendations_df['title'].tolist()
for i, title1 in enumerate(titles):
idx1 = df[df['title'] == title1].index[0]
for j, title2 in enumerate(titles[i+1:], i+1):
idx2 = df[df['title'] == title2].index[0]
similarity = cosine_similarity(tfidf_matrix[idx1:idx1+1],
tfidf_matrix[idx2:idx2+1])[0][0]
G.add_edge(title1, title2, weight=similarity)
# Visualization
pos = nx.spring_layout(G, k=1)
node_colors = ['#FF9999' if G.nodes[node]['type'] == 'Movie' else '#9999FF'
for node in G.nodes()]
node_sizes = [G.nodes[node]['score'] * 3000 for node in G.nodes()]
nx.draw(G, pos,
node_color=node_colors,
node_size=node_sizes,
with_labels=True,
font_size=8,
width=[G[u][v]['weight'] * 2 for u,v in G.edges()],
alpha=0.7)
plt.title('Recommendation Network')
return plt.gcf()
def get_recommendations(description):
recommendations_df = get_hybrid_recommendations(description)
bar_chart = create_bar_chart(recommendations_df)
network_graph = create_network_graph(recommendations_df)
return bar_chart, network_graph
# Create Gradio interface
iface = gr.Interface(
fn=get_recommendations,
inputs=gr.Textbox(label="Enter movie/series description", lines=3),
outputs=[
gr.Plot(label="Top 10 Recommendations"),
gr.Plot(label="Recommendation Network")
],
title="Netflix Recommendation System",
description="Enter a description of the content you're interested in to get personalized recommendations.",
theme="huggingface",
examples=[
["A thrilling action movie with car chases and explosions"],
["A romantic comedy about finding love in the city"],
["A documentary about nature and wildlife"]
]
)
if __name__ == "__main__":
iface.launch() |