About US

       京东科技从2017年开始建设智能运维,基于京东多年一线运维经验,以大数据和人工智能技术为抓手,形成以应用为中心的一体化智能运维解决方案。利用京东内部历年大促场景的数据积累,对算法进行不断的优化训练,在监控、数据库、网络、资源调度等多个纵向场景取得突破,可移植性强,自研通用化智能基线算法学件10+,自研通用化异常检测算法学件10+,场景化异常检测算法方案5+,具备多种自研通用化根因定位算法学件,可以自动触发多维实时根因定位 ,从上万维度属性值中定位到根因维度,自研5种以上增量式学习模板提取与相关分析算法学件,运维知识图谱内涵盖节点30W+,以应用为中心向外延伸出的图谱关系达90W+,赋能根因分析快速精准查询调用。发表IEEE国际会议论文(AIOps方向)9篇,申请50余项智能运维专利。

Team Members

Avatar

Xianbo Zhang

Software Architect

Avatar

Jing Zhang

Algorithm Engineer

Avatar

Qi Zhao

Software Developer Engineer

Avatar

Sheng Liu

Software Developer Engineer

Avatar

Xingxing Bu

Software Developer Engineer

Avatar

Ruize Xia

Software Developer Engineer

Recent Publications

Quickly discover relevant content by filtering publications.
(2022). PoSBert: Log Classification via Modified Bert Based on Part-of-Speech Weight. PRAI.

Cite DOI Link

(2022). Natural Language Processing-based Model for Log Anomaly Detection. SEAI.

Cite Link

(2022). CANet: Convolutional Attention-based Network for Multivariate Time Series Anomaly Detection. ICSP.

Cite Link

(2022). Multi-Attention Integrated Convolutional Network for Anomaly Detection of Time Series. ICCAE.

Cite Link

(2022). Threshold-free Anomaly Detection for Streaming Time Series through Deep Learning. ICMLA.

Cite Link

Awards

Contact

  • zhangjing511@jd.com
  • BeiJing Economic and Technological Development Zone Kechuang 11 Street
  • South Tower, 7th Floor, A block, Building 2, Headquarters of Jingdong Group