开发
一套基于
Python + Django 的鲜花推荐系统，可以分为以下几个核心部分：需求分析、技术架构设计、数据库建模、功能模块实现、推荐算法集成以及前端展示。下面我将详细为你讲解如何构建这样
一个系统。
---
###
一、系统目标
鲜花推荐系统的主要功能是根据用户的历史行为（如浏览、购买）、个人偏好、季节节日等因素，为用户智能推荐合适的鲜花产品。
---
### 二、技术栈
- **后端框架**：
Django
- **数据库**：PostgreSQL /
SQLite（开发阶段可用）
- **推荐算法**：协同过滤（User
-based 或 Item
-based）、基于内容的推荐、或简单的规则引擎
- **前端**：HTML/CSS/JS
+ Bootstrap，或使用
Django REST Framework
+ Vue/React
- **部署**：Gunicorn
+ Nginx
+ Docker（可选）
---
### 三、数据库模型设计（models.py）
```
python
from
django.db import models
from
django.contrib.auth.models import User
class Category
(models.Model
):
name = models.CharField
(max_length=100
)
description = models.TextField
(blank=True
)
def __str__
(self
):
return self.name
class Flower
(models.Model
):
name = models.CharField
(max_length=200
)
description = models.TextField
()
price = models.DecimalField
(max_digits=10, decimal_places=2
)
image = models.ImageField
(upload_
to='flowers/', blank=True, null=True
)
category = models.ForeignKey
(Category, on_delete=models.CASCADE
)
s
tock = models.PositiveIntegerField
(default=0
)
created_at = models.DateTimeField
(au
to_now_add=True
)
def __str__
(self
):
return self.name
class UserPreference
(models.Model
):
user = models.One
ToOneField
(User, on_delete=models.CASCADE
)
favorite_colors = models.CharField
(max_length=200, blank=True
) # e.g., "red,white,pink"
preferred_occasions = models.CharField
(max_length=200, blank=True
) # e.g., "birthday,anniversary"
def __str__
(self
):
return f"{self.user.username}'s preferences"
class PurchaseHis
tory
(models.Model
):
user = models.ForeignKey
(User, on_delete=models.CASCADE
)
flower = models.ForeignKey
(Flower, on_delete=models.CASCADE
)
purchased_at = models.DateTimeField
(au
to_now_add=True
)
quantity = models.PositiveIntegerField
()
class Meta:
ordering = ['
-purchased_at']
class ViewHis
tory
(models.Model
):
user = models.ForeignKey
(User, on_delete=models.CASCADE
)
flower = models.ForeignKey
(Flower, on_delete=models.CASCADE
)
viewed_at = models.DateTimeField
(au
to_now_add=True
)
class Meta:
ordering = ['
-viewed_at']
```
---
### 四、推荐逻辑实现（recommendations.py）
```
python
from collections import defaultdict
from .models import PurchaseHis
tory, ViewHis
tory, Flower, UserPreference
import ran
dom
def get_user_based_recommendations
(user,
top_n=5
):
"""
简化版用户协同过滤：找到相似用户买过但当前用户没买过的花。
"""
# 获取该用户已购买的花
purchased_flowers = PurchaseHis
tory.objects.filter
(user=user
).values_
list('flower_id', flat=True
)
# 获取所有其他用户的购买记录
all_purchases = PurchaseHis
tory.objects.exclude
(user=user
)
# 统计其他用户购买频率（按花统计）
flower_scores = defaultdict
(int
)
for purchase in all_purchases:
if purchase.flower_id not in purchased_flowers:
flower_scores[purchase.flower_id]
+= 1
recommended_ids = sorted
(flower_scores, key=flower_scores.get, reverse=True
)[:
top_n]
return Flower.objects.filter
(id__in=recommended_ids
)
def get_content_based_recommendations
(user,
top_n=5
):
"""
基于内容推荐：根据用户偏好颜色和场合推荐匹配的花。
这里假设花名或描述中包含关键词（实际应使用标签系统）。
"""
try:
pref = UserPreference.objects.get
(user=user
)
color_keywords = [c.strip
() for c in pref.favorite_colors.split
(','
) if c.strip
()]
occasion_keywords = [o.strip
() for o in pref.preferred_occasions.split
(','
) if o.strip
()]
except UserPreference.
DoesNotExist:
color_keywords = []
occasion_keywords = []
query = Flower.objects.all
()
scored_flowers = []
for flower in query:
score = 0
name_desc = f"{flower.name} {flower.description}".lower
()
for word in color_keywords:
if word.lower
() in name_desc:
score
+= 2
for word in occasion_keywords:
if word.lower
() in name_desc:
score
+= 3
if score > 0:
scored_flowers.
append
((flower, score
))
scored_flowers.sort
(key=lambda x: x[1], reverse=True
)
return [f[0] for f in scored_flowers[:
top_n]]
def hybrid_recommendations
(user,
top_n=5
):
"""
混合推荐：结合协同过滤与内容推荐
"""
cf_recs =
list(get_user_based_recommendations
(user,
top_n
))
cb_recs =
list(get_content_based_recommendations
(user,
top_n
))
# 合并并去重，保持顺序
seen = set
()
combined = []
for f in cf_recs
+ cb_recs:
if f.id not in seen:
combined.
append
(f
)
seen.add
(f.id
)
return combined[:
top_n]
```
---
### 五、视图函数示例（views.py）
```
python
from
django.shortcuts import render
from
django.contrib.auth.decora
tors import login_required
from .recommendations import hybrid_recommendations
@login_required
def recommended_flowers
(request
):
recommendations = hybrid_recommendations
(request.user,
top_n=8
)
return render
(request, 'flowers/recommended.html', {
'flowers': recommendations,
'title': '为您推荐'
}
)
```
---
### 六、URL 配置（urls.py）
```
python
from
django.urls import path
from . import views
urlpatterns = [
path
('recommendations/', views.recommended_flowers, name='recommendations'
),
]
```
---
### 七、模板示例（templates/flowers/recommended.html）
```html
<!
DOCTYPE html>
<html>
<head>
<title>{{ title }}</title>
<link href="https://cdn.jsdelivr.net/npm/bootstrap@5.3.0/dist/css/bootstrap.min.css" rel="stylesheet">
</head>
<body>
<div class="container mt
-4">
<h1>{{ title }}</h1>
<div class="row">
{% for flower in flowers %}
<div class="col
-md
-3 mb
-4">
<div class="card">
{% if flower.image %}
<img src="{{ flower.image.url }}" class="card
-img
-top" alt="{{ flower.name }}">
{% else %}
<img src="https://via.placeholder.com/300" class="card
-img
-top" alt="No Image">
{% endif %}
<div class="card
-body">
<h5 class="card
-title">{{ flower.name }}</h5>
<p class="card
-text">{{ flower.description|truncatewords:10 }}</p>
<p class="text
-primary"><strong>¥{{ flower.price }}</strong></p>
</div>
</div>
</div>
{% empty %}
<p>暂无推荐商品。</p>
{% endfor %}
</div>
</div>
</body>
</html>
```
---
### 八、解释说明
1. **模型设计**：
- `Flower` 表示鲜花信息；
- `UserPreference` 存储用户偏好（颜色、用途等）；
- `PurchaseHis
tory` 和 `ViewHis
tory` 记录用户行为，用于推荐。
2. **推荐算法**：
- 使用了简化版的 **协同过滤** 和 **基于内容的推荐**；
- 实际生产环境可替换为 Scikit
-learn、Surprise 库或 TensorFlow Recommenders。
3. **扩展性**：
- 可引入标签系统（Tag）对花进行更细粒度分类；
- 支持节日推荐（如母亲节主推康乃馨）；
- 加入评分系统（Rating）提升推荐精度。
4. **性能优化**：
- 推荐结果可缓存（Redis）；
- 异步任务（Celery）定时更新推荐列表。
---
###
本文分享了一个项目新手从爬虫转向Web开发的心路历程，并详细介绍了使用PyCharm和Django搭建项目的步骤，以及HTML、CSS和JavaScript等前端基础知识的学习资源。
扫一扫
308
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
