commonhas

Commonhas is a term used in information theory and data analysis to describe a framework for representing and locating shared content across multiple data streams through the use of hash-based signatures. The name combines "common" with "hash" to reflect its focus on identifying content that appears in more than one source.

Conceptually, commonhas relies on computing compact fingerprints for data segments and then aggregating these fingerprints in a central index. Each window or n-gram from each document is mapped to a fixed-size hash. When the same hash value appears across documents, those occurrences are flagged as candidates for further verification to determine whether they correspond to identical substrings or motifs.

Construction and parameters involve using a rolling hash or block-based hashing, with a specified n-gram size and sampling rate. The index stores for each hash value the set of documents and positions where it occurred. A subsequent filtering step compares candidate pairs to confirm actual equality, often using longer substrings or additional checks.

Applications include plagiarism detection, near-duplicate detection in web indexing, detection of reused code or content across repositories, and copyright enforcement. It can support scalable analysis by reducing comparisons to hash matches rather than all pairwise document comparisons.

Limitations include dependence on hash quality and chosen parameters, the potential for hash collisions requiring verification, the possibility of missing content that has undergone transformations, and the computational and storage costs that grow with the number of distinct fingerprints.

History notes that the term is not widely adopted as a standard technique but has appeared in theoretical discussions and experimental tooling in information retrieval and data deduplication literature during the 2010s and 2020s.

See also: Hash function, Rolling hash, MinHash, Shingling, Plagiarism detection, Data deduplication.