Data Quality & Proximinity Measurement

Question 1

Noise

Answer 1

Random and unpredictable variation in the data that is not related to the underlying pattern or signal that the model is trying to learn.

Question 2

Outliers

Answer 2

Data objects with characteristics that are considerably different than most of the other data objects in the data set.

Question 3

Missing values

Answer 3

The absence of a particular value in a dataset

Question 4

2 Reason for Missing Values

Answer 4

Information is not collected.
Attributes may not be applicable to all cases.

Question 5

4 Way to Handling Missing Values

Answer 5

Eliminate the attribute altogether - column

Eliminate instances/objects - row

Replace missing values.
Statistical methods - Median for age
Sophisticated methods - Regression imputation, KNN imputation

Prediction of missing values.

Question 6

Data Cleaning

Answer 6

Process of dealing with duplicate data issues.

Question 7

Proximity Measures

Answer 7

Mathematical metrics used to determine the similarity or distance between two data points in feature space.

Question 8

2 Distance measures used in clustering algorithms

Answer 8

K-means clustering
Hierarchical clustering,

Question 9

2 Distance measures used in classification algorithms

Answer 9

K-nearest neighbors (KNN)
Hierarchical Clustering.

Question 10

4 Way to Calculate Dissimilarity (Distance)

Answer 10

Manhattan Distance / Taxicab / City block / L1 norm
Euclidean Distance
Minkowski Distance
Mahalanobis Distance

Question 11

4 Way to Calculate Similarity Measurement

Answer 11

Simple Matching Coefficient (SMC)
Jaccard Coefficients
Cosine Similarity
Correlation
Drawback - Cannot detect non-linearity