Presented By O’Reilly and Intel AI
Put AI to work
April 10-11, 2018: Training
April 11-13, 2018: Tutorials & Conference
Beijing, CN

把AI注入BI: Kensho – 微软的自动化商业指标监控和分析工具

此演讲使用中文 (This will be presented in Chinese)

Tony Xing (Microsoft), Bixiong Xu (Microsoft)
14:5015:30 Friday, April 13, 2018
实施人工智能 (Implementing AI)
Location: 多功能厅5A+B(Function Room 5A+B) Level: Intermediate
Secondary topics:  设计AI平台(Designing AI platforms)

必要预备知识 (Prerequisite Knowledge)

对于机器学习的基本理解
A basic understanding of machine learning—both supervised and unsupervised learning

您将学到什么 (What you'll learn)

Kensho, 一个基于AI的商业指标监控与诊断工具, 我们通过将AI元素注入这个BI工具,从而构建来服务不同的微软团队的历程
Explore pain points from Microsoft customers and Kensho, Microsoft's automated business incident monitoring and diagnostics system

描述 (Description)

在微软,几乎每个团队都有一个通用的需求去监控产品和业务的指标,来了解他们的产品或者业务是不是处于正常状态。如果出现不符合历史趋势的异常,相关人员需要马上找到原因去解决问题。这个需求是简单和清晰的,但是构建一个令人满意的解决方案是非常困难的,首先是这个系统需要足够聪明能够自动从各种各样的指标的时间序列里学习到它们的正常模式,然后利用这个知识来智能的监控。第二,提供自动的洞察,为什么会发生这个问题,来帮助相关人员快速解决问题。

在这个议题里, 我们会概述这个领域内的用户痛点,我们会介绍Kensho, 一个基于AI的商业指标监控与诊断工具, 我们通过将AI元素注入这个BI工具,从而构建来服务不同的微软团队的历程。我们的从中学到的经验教训,技术的选择和演化,架构,算法等等。通过工程+数据科学解决了一个工业界的一个通用需求。另外我们会展望基于Kensho架构的下一代智能BI的解决方案。

内容

  • 用户痛点
  • 案例研究
  • 技术框架
  • 算法
  • 经验教训

At Microsoft, there is a common need to monitor product and business metrics to stay on top of product and business health and act quickly to address issues deviating from the historical pattern. The need is simple and clear. However, it turns out to be extremely difficult to have a scalable system that is also smart enough to provide automated anomaly detection across a wide spectrum of time series related to business or product performance as well as automated diagnostic insight into why that business incident happens.

Tony Xing and Bixiong Xu explore pain points from Microsoft customers and detail Microsoft’s path to implementing Kensho, an automated business incident monitoring and diagnostics system powered by AI to serve different Microsoft teams. Tony and Bixiong cover lessons learned during the process, the evolution of the technology, the solution architecture, and the algorithms and explain how collectively this combination of engineering and algorithms solve a common need that is applicable not only to Microsoft but to the industry as a whole.

Photo of Tony Xing

Tony Xing

Microsoft

Tony Xing is a senior product manager in AIDI (the AI, data, and infrastructure team) within Microsoft’s AI and Research organization. Previously, he was a senior product manager on the Skype data team within Microsoft’s Application and Service Group, where he worked on products for data ingestion, real-time data analytics, and the data quality platform.

Photo of Bixiong Xu

Bixiong Xu

Microsoft

Bixiong Xu is the principal dev manager on the AI, data, and infrastructure team at Microsoft.

Leave a Comment or Question

Help us make this conference the best it can be for you. Have questions you'd like this speaker to address? Suggestions for issues that deserve extra attention? Feedback that you'd like to share with the speaker and other attendees?

Join the conversation here (requires login)