Analytics

What is the nature of data mined without a clear outcome in mind?

Exploratory in nature, not pre-defined or classified, discovered from the data.

What are Association Rules used for?

They are used for discovering relationships among items.

What does X represent in the association rule X→Y?

X represents the antecedent (body) of the rule.

What does Y represent in the association rule X→Y?

Y represents the consequent (head) of the rule.

What does the Support Count measure?

The number of transactions containing the item-set.

What is the formula for calculating Support Percentage?

The formula is # of transactions with item set / total # of transactions.

What does Confidence in association rules indicate?

It indicates how often Y appears with transactions that contain X.

What is Lift in the context of association rules?

A measure of how much more likely two item sets co-occur.

What is the minimum support (minsupp) used for?

To specify a threshold below which item-sets are not considered frequent.

What is the key idea of the Apriori Algorithm?

If item set X is not frequent, then any item sets containing X cannot be frequent.

What does a high intra-similarity mean in clusters?

Points in the same cluster are similar to each other.

What does a lower inter-similarity indicate?

Points in different clusters are different from each other.

What is the purpose of Cluster Analysis?

To organize observations into meaningful clusters.

What is the significance of Distance Measures in clustering?

To assess similarity between individual data points.

What does Euclidean Distance measure?

The straight line distance between two points.

What is the first step in Data Normalization?

To eliminate specific units of measurement to transform them to a common scale.

What does Min-Max Normalization do?

Rescales attributes to have values between 0 and 1.

What is the role of Single Linkage in clustering?

It computes the minimum pairwise distance between points from different clusters.

What is Ward’s Method in clustering?

It minimizes within-cluster variance.

What is the process of Hierarchical Clustering?

It forms larger clusters from smaller ones.

What is a Dendrogram used for in clustering?

It shows the hierarchy of clusters.

What is Partitioning-based clustering?

It partitions data directly into K-groups, usually using Euclidean distance.

What does WSS stand for in assessing cluster quality?

Within Sum of Squared Errors.

What is indicated by a lower WSS?

More cohesive clusters with high intra-similarity.

What does BSS stand for in clustering analysis?

Between Sum of Squared Errors.

What does a higher BSS indicate?

A larger distance of centroids from each other, leading to lower inter-similarity.

How can the number of clusters be chosen using a Dendrogram?

By observing where clusters emerge naturally in the hierarchy.

What is an Elbow Plot used for?

To help decide the number of clusters by showing WSS/BSS as a function of the number of clusters.

What does it mean if Lift > 1?

Customers who buy X are more likely to buy Y.

What does it mean if Lift < 1?

Customers who buy X are less likely to buy Y.

What type of data does Matching Distance evaluate?

It evaluates binary data by counting the number of mismatches.

What does Standardization achieve?

Transforms each attribute to have a mean of 0 and standard deviation of 1.

What is the significance of computing cluster centroids in clustering?

To determine the average position of all points in a cluster.

What does Average Linkage involve?

It calculates the average pairwise distance between points from different clusters.

What does the process of recalculating cluster centroids in k-means aim to achieve?

To minimize the distance between points and their assigned centroids.

What happens during the ‘convergence’ step of k-means clustering?

Re-assignment of points doesn’t change anymore.

What does a good clustering algorithm aim to produce?

Clusters with low within-cluster variance and high between-cluster variance.

What role does domain knowledge play in defining minsupp and minconfidence?

It helps specify minimum acceptable thresholds based on business goals.

What can be observed in the layout of a Dendrogram?

The length of lines represents the proximity between clusters.

What is one limitation of hierarchical clustering?

It can be sensitive to outliers.