Unverified50% confidenceFactExact time
Time-series anomaly detection (TSAD) is a popular research direction at top conferences such as NeurIPS, SIGKDD, and VLDB.
1
Sources
50%
Confidence
Long-term
Relevance
8/31/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe TSB-AD-M benchmark, proposed by the Paparrizos team, is the standard dataset used by the vast majority of current TSAD papers to evaluate performance.64% similarUnverified行为检测技术通过监控API调用序列、注册表修改、网络连接等运行时行为来判断威胁，其原理可追溯至1990年代的异常检测理论60% similarUnverified变点检测（CPD）常用算法包括CUSUM、PELT和贝叶斯在线变点检测（BOCPD），用于识别时间序列数据中统计特性发生突变的时刻60% similarUnverifiedItanium uses speculative loads (the ld.s instruction) to execute memory reads early; triggered exceptions are deferred via the NaT bit and subsequently checked using the chk.s instruction.58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/829356API
curl https://kongchang.com/api/v1/knowledge/claims/829356MCP
get_claim(id=829356)