بهینه‌سازی دسته‌بندی خودکار پست‌ها با استفاده از شباهت متنی در جنگو Django

class PostCategory(models.Model):
    name = models.CharField(max_length=50)
    slug = models.SlugField(unique=True, blank=True, null=True, max_length=500)
    description = models.TextField(blank=True)
    is_active = models.BooleanField(default=True)
    created_at = models.DateTimeField(auto_now_add=True)
    updated_at = models.DateTimeField(auto_now=True)

    def __str__(self):
        return self.name

from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.metrics.pairwise import cosine_similarity
import numpy as np
from posts.models import PostCategory

class TextSimilarityService:
    def __init__(self):
        self.vectorizer = TfidfVectorizer(stop_words="english")
        self._update_category_vectors()

    def _update_category_vectors(self):
        categories = PostCategory.objects.all()
        category_texts = []
        self.categories = []

        for category in categories:
            category_text = f"{category.name} {category.description}"
            category_texts.append(category_text)
            self.categories.append(category)

        if category_texts:
            self.category_vectors = self.vectorizer.fit_transform(category_texts)
        else:
            self.category_vectors = None

    def find_similar_category(self, title: str, content: str, threshold: float = 0.1):
        default_category = PostCategory.objects.first()
        if not self.category_vectors or not title or not content:
            return default_category, 0.0

        combined_text = f"{title} {content}"
        combined_vector = self.vectorizer.transform([combined_text])
        similarities = cosine_similarity(combined_vector, self.category_vectors)
        best_match_index = np.argmax(similarities)
        best_match_score = similarities[0, best_match_index]

        if best_match_score >= threshold:
            return self.categories[best_match_index], best_match_score
        return default_category, 0.0

class Post(models.Model):
    title = models.CharField(max_length=200)
    content = models.TextField()
    category = models.ForeignKey(PostCategory, on_delete=models.CASCADE)
    created_at = models.DateTimeField(auto_now_add=True)

    def save(self, *args, **kwargs):
        if not self.category:
            similarity_service = TextSimilarityService()
            self.category, _ = similarity_service.find_similar_category(self.title, self.content)
        super().save(*args, **kwargs)