Cooccurente

Cooccurente is a term used in data analysis and information science to denote the simultaneous occurrence of two or more items within a defined window, such as a sentence, a time interval, or a spatial region. It is used to study the strength and significance of relationships between items when their joint presence carries meaning beyond individual occurrences.

Etymology: The word appears to be a constructed term formed from "co-occurrence" with the nominalizing suffix -ente, yielding a noun/adjective that can describe an event, pattern, or measure.

Applications: In linguistics, cooccurente refers to word pairs or larger collocations that appear together more often than expected by chance. In bioinformatics and epidemiology, it can describe simultaneous appearances of genes, phenotypes, or diseases within patient records or samples. In market research, it helps identify products or features that are frequently observed together.

Measurement and analysis: Data are organized into a cooccurrence matrix where rows and columns represent items, and cell values reflect joint counts. Common metrics include joint probability, pointwise mutual information (PMI), lift, and odds ratios. Statistical testing, such as chi-squared or permutation tests, can assess whether the observed cooccurrence exceeds random expectation.

Computation and interpretation: Analysts use cooccurente measures to infer associations, inform clustering, or support hypothesis generation. Limitations include sensitivity to sample size, window definition, and multiple testing.

See also: co-occurrence, collocation, mutual information, association rule learning.