Data Science Academy · 课时

从日期和文本构建特征

从原始字段中提取有效信号

从日期和文本构建特征 是 CoddyKit 上的免费 Data Science Academy 课时。这是第 4 节课，共 4 节。你可以在下方免费阅读本课时的完整内容 — 然后在浏览器中使用内置代码编辑器和全天候 AI 导师进行实践。这是 Data Science Academy 学习路径的一部分，你的进度在网页和 CoddyKit 应用中同步。 Data Science Academy 课程共包含 4 节课。

本课时的部分内容尚未翻译，以英文显示。

Raw Fields Hide Signal

A timestamp or a free-text note holds rich clues a model cannot read directly. Feature engineering pulls that signal into usable columns. 🔍

Parse the Date First

Before mining a date, convert the string with to_datetime. Only a real datetime exposes the handy parts you want to extract.

df['ts'] = pd.to_datetime(df['ts'])

The dt Accessor

The dt accessor unlocks date parts on a datetime column. Reach for it to grab year, month, day, and more in one line.

df['year'] = df['ts'].dt.year
df['month'] = df['ts'].dt.month

Day of Week

Pull dayofweek to capture weekly rhythm, where Monday is 0 and Sunday is 6. Behavior often shifts a lot by weekday.

df['weekday'] = df['ts'].dt.dayofweek

Weekend Flag

Turn the weekday into a simple boolean: is this a weekend? Such yes-or-no flags are easy for models to learn from.

df['is_weekend'] = df['ts'].dt.dayofweek >= 5

Cyclical Time

Hour 23 sits right next to hour 0, yet the numbers look far apart. Cyclical encoding with sine and cosine fixes that wrap-around.

Text: Length and Counts

Start simple with text. The character length of a review or its word count is often a surprisingly strong feature.

df['len'] = df['review'].str.len()
df['words'] = df['review'].str.split().str.len()

Contains a Keyword

Flag whether text holds a key term using str.contains. A column for words like refund can capture clear intent.

df['has_refund'] = df['review'].str.contains('refund')

Bag of Words

To use words themselves, count them. A bag-of-words turns text into one column per term holding how often it appears.

from sklearn.feature_extraction.text import CountVectorizer
X = CountVectorizer().fit_transform(df['review'])

TF-IDF Weighs Words

Plain counts overrate common words. TF-IDF rewards terms that are frequent in one row yet rare overall, so the distinctive words stand out.

from sklearn.feature_extraction.text import TfidfVectorizer
X = TfidfVectorizer().fit_transform(df['review'])

Avoid Leaky Features

Never build a feature from information unavailable at prediction time. A leaky column inflates scores in testing, then fails in the real world.

Quick Check

You have a datetime column and want the weekday number. What do you use?

Recap: Date and Text Features

You mined raw fields into signal: parse dates then use dt for parts and flags, and turn text into lengths, keyword flags, or word counts. 🎉

常见问题解答

「从日期和文本构建特征」课时是免费的吗？

是的 — 「从日期和文本构建特征」的完整文本可在网页上免费阅读。要进行交互式练习（内置代码编辑器和全天候 AI 导师）并解锁 Data Science Academy 课程的其余内容，请升级到 CoddyKit PRO。 Data Science Academy 课程共包含 4 节课。

「从日期和文本构建特征」这节课中我会学到什么？

从原始字段中提取有效信号你通过在浏览器中直接运行的动手代码来练习 Data Science Academy，全天候 AI 导师会在你学习这节课的过程中回答你的问题。

学习 Data Science Academy 需要有经验吗？

无需任何先前经验。CoddyKit 上的 Data Science Academy 课程适合初学者到高级学习者，你可以从这里开始或从头开始，按照自己的节奏学习。这是第 4 节课，共 4 节。

「从日期和文本构建特征」课时需要多长时间？

大多数 CoddyKit 课程大约需要 5–10 分钟。每节课都很精短且互动，所以你能稳步进步，并在网页和应用中从离开的地方继续。

我能在这节 Data Science Academy 课中编写并运行代码吗？

能。每节 Data Science Academy 课都包含内置代码编辑器，你可以在浏览器中直接编写并运行真实代码，并获得即时 AI 反馈 — 无需本地设置。

此课程中的所有课时

← 返回 Data Science Academy