Regular expressions for text cleaning in NLP - Model Metrics & Evaluation

Metrics & Evaluation - Regular expressions for text cleaning

Which metric matters for this concept and WHY

When using regular expressions for text cleaning, the key metric is data quality improvement. This means how well the cleaning removes unwanted characters or patterns without losing important information. Metrics like precision and recall can be adapted here:

• Precision: How many of the removed parts were actually unwanted (true removals vs wrong removals).
• Recall: How many unwanted parts were successfully removed (true removals vs missed unwanted parts).

Good cleaning keeps useful text intact (high precision) and removes all noise (high recall). This helps models learn better from clean data.

Confusion matrix or equivalent visualization (ASCII)

Unwanted parts:  | Removed | Not Removed
-----------------|---------|------------
Actually Removed |   TP    |    FN      
Actually Kept    |   FP    |    TN      

TP = Unwanted parts correctly removed
FP = Useful parts wrongly removed
FN = Unwanted parts missed
TN = Useful parts kept
    

Precision vs Recall tradeoff with concrete examples

Imagine cleaning tweets:

• High precision, low recall: You only remove very obvious noise like URLs, but miss emojis or hashtags. You keep most useful text but some noise remains.
• High recall, low precision: You remove all special characters including some words or emojis that carry meaning. You get rid of noise but lose useful info.

Balance is key: remove enough noise to help the model but keep important text for learning.

What "good" vs "bad" metric values look like for this use case

• Good cleaning: Precision > 0.9 and Recall > 0.85. Most noise removed, very few useful parts lost.
• Bad cleaning: Precision < 0.7 or Recall < 0.5. Either too much useful text removed or too much noise left.

Good cleaning leads to better model accuracy and faster training.

Metrics pitfalls

• Accuracy paradox: Simply counting how many characters removed is misleading. Removing too much can look like high "accuracy" but harms data quality.
• Data leakage: Over-cleaning can remove important signals that models need, causing poor generalization.
• Overfitting indicators: If cleaning is too strict on training data patterns, model may fail on new text with different noise.

Self-check question

Your text cleaning removes 98% of unwanted noise but also removes 30% of useful words (low precision). Is this good?

Answer: No, because losing 30% of useful words means the model will miss important information. You should improve precision to keep useful text while still removing noise.

Key Result

Effective text cleaning balances high precision and recall to remove noise while preserving useful text.

Practice

1. What is the main purpose of using regular expressions in text cleaning for NLP?

easy

A. To find and remove unwanted patterns or characters in text

B. To train machine learning models directly

C. To store large datasets efficiently

D. To visualize text data with graphs

Solution

1. Step 1: Understand the role of regular expressions

   Regular expressions are used to identify patterns in text, such as unwanted characters or specific sequences.

2. Step 2: Connect to text cleaning

   Text cleaning involves removing or replacing unwanted parts of text to prepare it for analysis or modeling.

3. Final Answer:

   To find and remove unwanted patterns or characters in text -> Option A

4. Quick Check:

   Regular expressions clean text by pattern matching [OK]

Hint: Regular expressions = pattern search and replace in text [OK]

Common Mistakes:

• Confusing regex with model training
• Thinking regex stores data
• Assuming regex creates visualizations

2. Which of the following is the correct Python syntax to import the regular expression module?

easy

A. from regex import *

B. import regex

C. import re

D. import regular_expression

Solution

1. Step 1: Recall Python's regex module name

   Python's built-in module for regular expressions is named 're'.

2. Step 2: Check syntax correctness

   The correct import statement is 'import re' to use regex functions.

3. Final Answer:

   import re -> Option C

4. Quick Check:

   Python regex module = re [OK]

Hint: Remember: Python regex module is 're' not 'regex' [OK]

Common Mistakes:

• Using 'import regex' which is not standard
• Trying to import non-existent modules
• Confusing module names with function names

3. What will be the output of this Python code snippet?

import re
text = "Hello, World! 123"
cleaned = re.sub(r'[^a-zA-Z ]', '', text)
print(cleaned)

medium

A. Hello World

B. Hello World 123

C. Hello, World!

D. HelloWorld123

Solution

1. Step 1: Understand the regex pattern used

   The pattern '[^a-zA-Z ]' means any character NOT a letter (a-z or A-Z) or space.

2. Step 2: Apply re.sub to remove unwanted characters

   All characters except letters and spaces are removed, so commas, exclamation marks, and digits are deleted.

3. Final Answer:

   Hello World -> Option A

4. Quick Check:

   Regex removes non-letters/spaces = 'Hello World ' [OK]

Hint: [^...] means NOT those characters, so it removes digits and punctuation [OK]

Common Mistakes:

• Thinking digits remain after substitution
• Confusing character classes with ranges
• Ignoring spaces in the pattern

4. Identify the error in this regex code snippet for removing digits from text:

import re
text = "Price: 100 dollars"
cleaned = re.sub(r'\d', '', text)
print(cleaned)

medium

A. The pattern '\d' should be '\D' to remove digits

B. The backslash in '\d' is not escaped properly

C. The re.sub function is used incorrectly

D. The code will run correctly and remove digits

Solution

1. Step 1: Check regex pattern correctness

   The pattern r'\d' correctly matches digits (0-9).

2. Step 2: Verify code syntax and function usage

   The code uses raw string r'\d' which properly escapes the backslash, so digits are removed as intended.

3. Final Answer:

   The code will run correctly and remove digits -> Option D

4. Quick Check:

   r'\d' matches digits; re.sub removes them correctly [OK]

Hint: In raw strings, r'\d' matches digits; no extra escaping needed [OK]

Common Mistakes:

• Thinking '\d' needs double escaping outside raw strings
• Confusing '\d' with '\D' (non-digit)
• Assuming re.sub syntax is wrong

5. You want to clean a text dataset by removing all URLs and extra spaces. Which regex pattern and code snippet correctly achieves this in Python?

import re
text = "Visit https://example.com now!  Enjoy!"
cleaned = re.sub(_____, ' ', text)
cleaned = re.sub(r'\s+', ' ', cleaned).strip()
print(cleaned)

hard

A. r'http://[a-z]+'

B. r'https?://\S+'

C. r'www\.[a-z]+\.com'

D. r'https?://[a-z]+'

Solution

1. Step 1: Identify a regex pattern that matches URLs

   The pattern 'https?://' matches 'http://' or 'https://', and '\S+' matches non-space characters following it, capturing full URLs.

2. Step 2: Understand the code's cleaning steps

   First, URLs are replaced by a space, then multiple spaces are reduced to one, and leading/trailing spaces removed.

3. Final Answer:

   r'https?://\S+' -> Option B

4. Quick Check:

   Use 'https?://\S+' to remove URLs effectively [OK]

Hint: Use 'https?://' plus '\S+' to match full URLs [OK]

Common Mistakes:

• Using too narrow patterns missing https or full URL
• Not removing extra spaces after substitution
• Using patterns that match only partial URLs