Files
simulated-examination/q2_1_crawler/数据分析.py

139 lines
3.5 KiB
Python
Raw Permalink Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

# 直接在代码里写死数据,不依赖外部文件
movies = [
{
"id": 1,
"title": "电影A",
"director": "导演A",
"year": 2021,
"rating": 9.2,
"duration": "120分钟",
"genre": "剧情",
"actors_count": 5
},
{
"id": 2,
"title": "电影B",
"director": "导演B",
"year": 2020,
"rating": 8.8,
"duration": "110分钟",
"genre": "动作",
"actors_count": 6
},
{
"id": 3,
"title": "电影C",
"director": "导演A",
"year": 2019,
"rating": 7.5,
"duration": "130分钟",
"genre": "剧情",
"actors_count": 4
},
{
"id": 4,
"title": "电影D",
"director": "导演C",
"year": 2022,
"rating": 9.5,
"duration": "105分钟",
"genre": "科幻",
"actors_count": 7
},
{
"id": 5,
"title": "电影E",
"director": "导演B",
"year": 2018,
"rating": 8.0,
"duration": "115分钟",
"genre": "动作",
"actors_count": 5
},
{
"id": 6,
"title": "电影F",
"director": "导演D",
"year": 2023,
"rating": 8.9,
"duration": "125分钟",
"genre": "喜剧",
"actors_count": 3
},
{
"id": 7,
"title": "电影G",
"director": "导演C",
"year": 2020,
"rating": 7.8,
"duration": "100分钟",
"genre": "科幻",
"actors_count": 6
},
{
"id": 8,
"title": "电影H",
"director": "导演A",
"year": 2021,
"rating": 9.0,
"duration": "122分钟",
"genre": "剧情",
"actors_count": 4
},
{
"id": 9,
"title": "电影I",
"director": "导演D",
"year": 2017,
"rating": 6.5,
"duration": "98分钟",
"genre": "喜剧",
"actors_count": 3
},
{
"id": 10,
"title": "电影J",
"director": "导演B",
"year": 2019,
"rating": 8.3,
"duration": "118分钟",
"genre": "悬疑",
"actors_count": 5
}
]
# 1. 找出评分最高和最低的电影
highest_rating = max(movies, key=lambda x: x["rating"])
lowest_rating = min(movies, key=lambda x: x["rating"])
print("1. 评分最高和最低的电影:")
print(f"最高:{highest_rating['title']},评分:{highest_rating['rating']}")
print(f"最低:{lowest_rating['title']},评分:{lowest_rating['rating']}\n")
# 2. 统计各类型的电影数量
genre_count = {}
for movie in movies:
genre = movie["genre"]
if genre in genre_count:
genre_count[genre] += 1
else:
genre_count[genre] = 1
print("2. 各类型电影数量:")
print(genre_count, "\n")
# 3. 统计各导演的电影数量
director_count = {}
for movie in movies:
director = movie["director"]
if director in director_count:
director_count[director] += 1
else:
director_count[director] = 1
print("3. 各导演电影数量:")
print(director_count, "\n")
# 4. 统计2020年以后上映的电影数量
count_2020 = 0
for movie in movies:
if movie["year"] >= 2020:
count_2020 += 1
print(f"4. 2020年以后上映的电影数量{count_2020}")