Cluster analysis

Cluster analysis or clustering is the task of grouping a set of objects in such a way that objects in the same group (called a cluster) are more <a href="/facts/Similarity_measure/FmmFco8c">similar</a> (in some specific sense defined by the analyst) to each other than to those in other groups (clusters). It is a main task of <a href="/facts/Exploratory_data_analysis/8hvnf7JS">exploratory data analysis</a>, and a common technique for <a href="/facts/Statistics/DNPsKGYU">statistical</a> <a href="/facts/Data_analysis/CNtARxLW">data analysis</a>, used in many fields, including <a href="/facts/Pattern_recognition/o1s38opj">pattern recognition</a>, <a href="/facts/Image_analysis/woLB8cX7">image analysis</a>, <a href="/facts/Information_retrieval/MFSmpeNw">information retrieval</a>, <a href="/facts/Bioinformatics/D5x2L8ee">bioinformatics</a>, <a href="/facts/Data_compression/gMFuY4FT">data compression</a>, <a href="/facts/Computer_graphics/lapBE8wv">computer graphics</a> and <a href="/facts/Machine_learning/e0w0XJTu">machine learning</a>.
Cluster analysis refers to a family of algorithms and tasks rather than one specific <a href="/facts/Algorithm/fnl5NmRt">algorithm</a>. It can be achieved by various algorithms that differ significantly in their understanding of what constitutes a cluster and how to efficiently find them. Popular notions of clusters include groups with small <a href="/facts/Distance_function/RkUptSo8">distances</a> between cluster members, dense areas of the data space, intervals or particular <a href="/facts/Statistical_distribution/frWIF9F9">statistical distributions</a>. Clustering can therefore be formulated as a <a href="/facts/Multi-objective_optimization/1Hb4gjGJ">multi-objective optimization</a> problem. The appropriate clustering algorithm and parameter settings (including parameters such as the <a href="/facts/Metric_(mathematics)/RkUptSo8">distance function</a> to use, a density threshold or the number of expected clusters) depend on the individual <a href="/facts/Data_set/eDvegUTK">data set</a> and intended use of the results. Cluster analysis as such is not an automatic task, but an iterative process of <a href="/facts/Knowledge_discovery/eopEv228">knowledge discovery</a> or interactive multi-objective optimization that involves trial and failure. It is often necessary to modify <a href="/facts/Data_preprocessing/CS2W8m5B">data preprocessing</a> and model parameters until the result achieves the desired properties.
Besides the term clustering, there are a number of terms with similar meanings, including automatic <a href="/facts/Statistical_classification/jXXHRkXR">classification</a>, <a href="/facts/Numerical_taxonomy/ctS4h4XA">numerical taxonomy</a>, botryology (from <a href="/facts/Greek_language/pzDd6eIz">Greek</a>: βότρυς 'grape'), typological analysis, and <a href="/facts/Community_structure/CKfUKFYh">community detection</a>. The subtle differences are often in the use of the results: while in data mining, the resulting groups are the matter of interest, in automatic classification the resulting discriminative power is of interest.
Cluster analysis originated in anthropology by Driver and Kroeber in 1932 and introduced to psychology by <a href="/facts/Joseph_Zubin/Q4uyTUks">Joseph Zubin</a> in 1938 and <a href="/facts/Robert_Tryon/1Pjsvazg">Robert Tryon</a> in 1939 and famously used by <a href="/facts/Raymond_Cattell/xSH3BLI3">Cattell</a> beginning in 1943 for trait theory classification in <a href="/facts/Personality_psychology/AWhpI44I">personality psychology</a>.

Cluster analysis open-in-new

Cluster analysis