Relevantti

Relevanti is a conceptual framework in information science for evaluating and quantifying the relevance of data items to user queries or tasks. It is used to guide ranking, filtering, and summarization in information systems.

The core idea is to represent relevance as a score produced by a scoring function that combines contextual factors (user intent, domain, time, provenance) with intrinsic item properties (content features, metadata, popularity).

Signals include textual similarity, recency, diversity, novelty, and user feedback. The framework supports multi-criteria scoring and explainable outputs.

Implementation often involves machine learning models that learn to adjust weights from interaction data; it emphasizes context awareness and adaptability to user preferences.

Applications include search engines, recommendation systems, knowledge management, and content curation.

Criticism centers on the subjectivity of relevance, context dependence, and potential biases in training data, as well as the computational cost of complex scoring.

Notes: The term "relevantti" appears as a neologism formed from "relevant" and is used in some discussions of Finnish-language information systems.