BT

Facilitating the Spread of Knowledge and Innovation in Professional Software Development

Write for InfoQ

Topics

Choose your language

InfoQ Homepage Guides The InfoQ eMag: Real-World Machine Learning: Case Studies, Techniques and Risks

The InfoQ eMag: Real-World Machine Learning: Case Studies, Techniques and Risks

Bookmarks

Machine learning (ML) and deep-learning technologies like Apache Spark, Flink, Microsoft CNTK, TensorFlow, and Caffe brought data analytics to the developer community. Whether it's classifying 2 million sales products received from over 700 multinational retailers for the "Love the Sales" website, building awareness of hindsight bias with customers at Salesforce Einstein, or understanding software system behavior with machine learning and time-series data at Sumo Logic, machine-learning solutions are driving the competitive innovation edge in companies and industries.

This eMag focuses on the current landscape of machine-learning technologies and real-world case studies of applied machine learning.

Free download

The InfoQ eMag: Real-World Machine Learning: Case Studies, Techniques and Risks includes:

  • Get More Bytes for Your BuckLovethesales.com had to classify one million product data from 700 different disparate sources across a large domain. They decided to create a hierarchy of classifiers through utilizing machine learning, specifically Support Vector Machines.  They learned that optimising the way in which the svms were connected together yielded vast improvements in the reuse of labeled training data.
  • Back to the Future: Demystifying Hindsight Bias - Enterprise AI has more prevalent nuances in the input data than in consumer AI or academia. The Achilles’ heel in this domain is Hindsight Bias. In layman terms, it is like Marty McFly (from Back to the Future) traveling to the future, getting his hands on the Sports Almanac, and using it to bet on the games of the present. Mayukh Bhaowal from Salesforce Einstein explains how to counteract it.
  • Understanding Software System Behaviour With ML and Time Series Data - David Andrzejewski presented "Understanding Software System behaviour With ML and Time Series Data".  This article is a summary of his presentation and an overview on what to look out for. Know about the traditional approaches to time series, how to handle missing values, and know about possibly occurring seasonality in your data. Be careful about what threshold you set for anomaly detection.
  • Analysing & Preventing Unconscious Bias in Machine Learning - This article is based on Rachel Thomas’s keynote presentation, “Analysing & Preventing Unconscious Bias in Machine Learning” at QCon.ai 2018. She talks about three case studies, attempting to diagnose bias, identify some sources, and discusses what it takes to avoid it.
  • Can People Trust the Automated Decisions Made by Algorithms? - The use of automated decision making is increasing. These algorithms can produce results that are incomprehensible, or socially undesirable. How can we determine the safety of algorithms in devices if we cannot understand them? Public fears about the inability to foresee adverse consequences has impeded technologies such as nuclear energy and genetically modified crops.

InfoQ eMags are professionally designed, downloadable collections of popular InfoQ content - articles, interviews, presentations, and research - covering the latest software development technologies, trends, and topics.

BT