--- title: "Clustering, need and qualities" subject: "Data Mining" chapter: 5 key: "semester_7/data_mining:clustering-define-qualities" source: https://www.bctengineer.com/q/semester_7/data_mining/clustering-define-qualities site: BCT Engineering --- # Clustering, need and qualities Data Mining · Chapter 5: Cluster Analysis - Asked in: year not given - Video: [Lec - 22: Clustering in Data Mining Explained | Top Clustering Methods You MUST Know!](https://www.youtube.com/watch?v=c8yiJq1do38) — Gate Smashers - Answer checked by Claude Opus 5.5 (high effort) on 2026-10-14 - Page: https://www.bctengineer.com/q/semester_7/data_mining/clustering-define-qualities - Whole chapter: https://www.bctengineer.com/md/semester_7/data_mining/chapter-5 ## Question Define Clustering. Why do we need cluster analysis? Discuss the qualities of a good cluster. ## Short answer Clustering groups unlabeld objects into clusters with high intra-cluster similarity and low inter-cluster similarity. It discovers natural structure, compresses data, and detects outliers. Good clusters and methods are compact, well-separated, scalable, can handle arbitrary shapes, noise, high dimensionality, and produce interpretable results. ## Answer **Clustering** is the process of grouping a set of data objects into multiple groups (clusters) so that objects within a cluster have **high similarity** to each other and are **dissimilar** to objects in other clusters. It is unsupervised (learning by observation): no class labels are given. *Example*: grouping customers of a shop by their purchase pattern. #### Why cluster analysis is needed - Discover natural groups and the structure of unlabelled data. - Summarise and compress large data sets with cluster representatives. - Find outliers (e.g. fraud) and prepare data for other methods. - Applications: market segmentation, document grouping, image segmentation, gene analysis, city planning. #### Qualities of a good cluster / clustering method 1. **High intra-cluster similarity** (compactness) and **low inter-cluster similarity** (good separation). 2. **Scalability**: works for small and very large data. 3. Ability to handle **different attribute types** (numeric, categorical, mixed, binary). 4. Discovery of clusters with **arbitrary shape**, not only spherical ones. 5. **Minimal domain knowledge** required to fix input parameters (such as k). 6. Ability to deal with **noisy data, outliers and missing values**. 7. **Insensitivity to the order** of input records. 8. Handling **high dimensionality**. 9. Incorporation of **user constraints**. 10. **Interpretability and usability**: the clusters should be meaningful. The quality is measured by a good similarity measure and by indices such as SSE, silhouette coefficient and cohesion/separation. ## Check yourself The answer's key points as multiple-choice questions, with the correct option and why. 1. Which best defines clustering? A. Assigning objects to predefined classes using labelled examples B. Grouping objects so that those in a cluster are highly similar and dissimilar to those in other clusters C. Finding rules that link items bought together D. Removing duplicate records Answer: B. Clustering is unsupervised grouping with high intra-cluster and low inter-cluster similarity. (Key point: Definition of clustering) 2. A good clustering has which combination? A. Low intra-cluster and high inter-cluster similarity B. High intra-cluster and high inter-cluster similarity C. High intra-cluster and low inter-cluster similarity D. Low similarity everywhere Answer: C. Compact clusters (high intra similarity) that are well separated (low inter similarity) mark good quality. (Key point: Good cluster criterion) 3. Why is 'discovery of clusters with arbitrary shape' a desired quality? A. Many methods find only spherical clusters, but real clusters can have any shape B. Spherical clusters are never found in data C. Shapes decide the number of attributes D. It removes the need for a distance measure Answer: A. Distance-based methods like k-means tend toward spherical clusters, so handling arbitrary shape is a quality criterion. (Key point: Arbitrary shape) 4. What does 'minimal domain knowledge to fix input parameters' mean? A. The user must know the true classes B. The method should not force the user to guess parameters such as k C. Only domain experts may run it D. Input data must be preprocessed manually Answer: B. A good method should not depend heavily on user-specified parameters such as the number of clusters. (Key point: Minimal domain knowledge) ## Related in Data Mining - [Need of clustering; hierarchical vs partitioning; K-means](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Acluster-need-hier-vs-partition-kmeans) — Data Mining, chapter 5 - [Cluster analysis; K-means on height/width](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Akmeans-height-width) — Data Mining, chapter 5 - [Cluster analysis applications; K-means](https://www.bctengineer.com/question.md?key=semester_7%2Fdata_mining%3Acluster-application-kmeans) — Data Mining, chapter 5 ## Related in other subjects - [Leadership and qualities of a good leader](https://www.bctengineer.com/question.md?key=semester_7%2Forganization_and_management%3Aleadership-good-leader-qualities) — Organization and Management, chapter 8, asked 2× - [Why do we need DFT](https://www.bctengineer.com/question.md?key=semester_7%2Fdigital_signal_analysis_and_processing%3Awhy-need-dft) — Digital Signal Analysis and Processing, chapter 7, asked 6× - [Why do we need FFT](https://www.bctengineer.com/question.md?key=semester_7%2Fdigital_signal_analysis_and_processing%3Awhy-need-fft) — Digital Signal Analysis and Processing, chapter 7, asked 2× - [Need of analysis and design](https://www.bctengineer.com/question.md?key=semester_1%2Fcomputer_programming%3Aanalysis-design-before-coding) — Computer Programming, chapter 2