Unverified50% confidenceFactExact time
Category purity is a commonly used metric in cluster analysis: for each sample cluster corresponding to a category in the latent space, compute the proportion of samples belonging to the majority class within the cluster, then average over all clusters.
1
Sources
50%
Confidence
Long-term
Relevance
11/17/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedA class prototype is the mean of all sample feature vectors of a given class in feature space, with low storage overhead and no leakage of raw data.68% similarUnverified分层采样(stratified sampling)先将用户查询按意图类型分类或通过embedding聚类,再从每个类别按比例或等量采样,比纯随机采样更能确保评估集代表性65% similarUnverifiedThe score type defines each scoring level through an ordered list of criteria, suitable for measuring a certain attribute by grade.65% similarUnverifiedFor real-valued responses, a family of intervals can replace the traditional scalar thresholds; for categorical responses, a family of singleton sets fully determines the distribution.64% similarUnverifiedTyped centrality means building a separate graph or normalizing separately for each class of nodes before computing centrality, so that importance evaluation is comparable within the same semantic category.62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1179419API
curl https://kongchang.com/api/v1/knowledge/claims/1179419MCP
get_claim(id=1179419)