Error classification in GraphQL - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is error classification in data science?

Error classification is the process of identifying and categorizing different types of errors that occur in data analysis or machine learning models, such as false positives, false negatives, or misclassifications.

beginner

What is a false positive error?

A false positive error happens when a model incorrectly predicts a positive outcome for a case that is actually negative. For example, a spam filter marking a normal email as spam.

beginner

What is a false negative error?

A false negative error happens when a model incorrectly predicts a negative outcome for a case that is actually positive. For example, a medical test missing a disease that is present.

intermediate

Why is it important to classify errors in machine learning?

Classifying errors helps us understand where a model makes mistakes, so we can improve it. It also helps us decide which errors are more costly and need more attention.

intermediate

What is the difference between accuracy and error classification?

Accuracy measures how often a model is correct overall, while error classification breaks down the types of mistakes the model makes, like false positives and false negatives.

Which of the following is a false positive error?

A Predicting normal for a spam email

B Predicting spam for a normal email

C Predicting disease when no disease is present

D Predicting no disease when disease is present

A false positive is when the model predicts a positive outcome incorrectly, like marking a normal email as spam or predicting disease when no disease is present.

What does a false negative error mean?

A Model predicts negative but actual is positive

B Model predicts positive but actual is negative

C Model predicts positive and actual is positive

D Model predicts negative and actual is negative

A false negative means the model missed a positive case by predicting negative.

Why do we classify errors in machine learning?

A To understand and improve model performance

B To ignore mistakes

C To increase the number of errors

D To make the model slower

Classifying errors helps us understand where the model fails and how to improve it.

Which error type is more critical in a medical diagnosis model?

A False positive

B False negative

C True positive

D True negative

False negatives are critical because missing a disease can have serious consequences.

Accuracy tells us:

A Only false positives

B Types of errors the model makes

C How often the model is correct overall

D Only false negatives

Accuracy measures the overall correctness of the model's predictions.

Explain the difference between false positive and false negative errors with simple examples.

Think about spam emails and medical tests.

Why is error classification important when evaluating a machine learning model?

Consider how different errors affect decisions.

Practice

1. What is the main purpose of error classification in data science models?

easy

A. To group model mistakes into clear categories for better understanding

B. To increase the speed of the model's predictions

C. To reduce the size of the dataset

D. To improve the model's user interface

Solution

1. Step 1: Understand error classification

   Error classification means sorting mistakes into groups to see patterns.

2. Step 2: Identify the goal of grouping errors

   Grouping helps find where the model works well or needs improvement.

3. Final Answer:

   To group model mistakes into clear categories for better understanding -> Option A

4. Quick Check:

   Error classification = Group mistakes [OK]

Hint: Error classification groups mistakes to spot issues fast [OK]

Common Mistakes:

• Thinking error classification speeds up predictions
• Confusing error classification with data size reduction
• Believing it improves user interface

2. Which GraphQL query structure correctly requests error counts grouped by error type?

easy

A. query { errorCountList { errorType, total } }

B. query { errors { count type } }

C. query { errorCounts(type, count) }

D. query { errorCounts { type count } }

Solution

1. Step 1: Identify correct GraphQL syntax

   GraphQL queries use curly braces with fields inside, like { errorCounts { type count } }.

2. Step 2: Check field names and order

   query { errorCounts { type count } } uses correct field names and structure matching typical GraphQL style.

3. Final Answer:

   query { errorCounts { type count } } -> Option D

4. Quick Check:

   Correct GraphQL query = query { errorCounts { type count } } [OK]

Hint: GraphQL queries list fields inside braces in correct order [OK]

Common Mistakes:

• Using parentheses instead of braces
• Swapping field names order incorrectly
• Using wrong field names not defined in schema

3. Given this GraphQL query:

query { errorCounts { type count } }

And this response:

{ "data": { "errorCounts": [ { "type": "False Positive", "count": 5 }, { "type": "False Negative", "count": 3 } ] } }

What is the total number of errors reported?

medium

A. 3

B. 8

C. 5

D. 15

Solution

1. Step 1: Extract counts from response

   The response shows 5 False Positives and 3 False Negatives.

2. Step 2: Add the counts together

   5 + 3 = 8 total errors.

3. Final Answer:

   8 -> Option B

4. Quick Check:

   5 + 3 = 8 total errors [OK]

Hint: Add all error counts from response to get total [OK]

Common Mistakes:

• Choosing only one error type count
• Multiplying counts instead of adding
• Ignoring one of the error types

4. This GraphQL query returns an error:

query { errorCounts { type, count } }

What is the likely cause?

medium

A. Incorrect field names

B. Missing query keyword

C. Using commas between fields inside braces

D. Using parentheses instead of braces

Solution

1. Step 1: Check GraphQL field syntax

   GraphQL fields inside braces should not have commas separating them.

2. Step 2: Identify the error cause

   The query uses commas between fields, causing syntax error.

3. Final Answer:

   Using commas between fields inside braces -> Option C

4. Quick Check:

   No commas between fields in GraphQL [OK]

Hint: Remove commas between fields in GraphQL queries [OK]

Common Mistakes:

• Adding commas between fields
• Omitting the query keyword (optional but valid)
• Using wrong field names without checking schema

5. You want to find which error type has the highest count using GraphQL. Which query and approach is best?

hard

A. query { errorCounts { type count } } and then find max count in client code

B. query { maxErrorCount } to get highest count directly

C. query { errorCounts(type: "max") { count } } to filter max error

D. query { errorCounts { count } } and assume first is max

Solution

1. Step 1: Understand GraphQL query capabilities

   GraphQL typically returns lists; aggregation like max often done client-side.

2. Step 2: Choose approach to find max error count

   Query all error counts and find max in client code is standard and flexible.

3. Final Answer:

   query { errorCounts { type count } } and then find max count in client code -> Option A

4. Quick Check:

   Get all counts, find max outside GraphQL [OK]

Hint: Fetch all counts, then find max in your code [OK]

Common Mistakes:

• Expecting GraphQL to do max aggregation automatically
• Filtering by 'max' as a type which is invalid
• Assuming first item is max without checking