Machine Learning

Eradium dictionary machine learning

Machine Learning

Machine learning (ML) is the subfield of computer science that “gives computers the ability to learn without being explicitly programmed” (1). To apply Machine learning we need a set of training data and an algorithm that capable to learn from the data. The training data can be just a ready stream of data, set of images, audio or video signals that converted to digital format and provided as the training set.

For instance, we can provide hundreds or images with black and white cats mixing with any other photos and train computer to find all white and black cats in any new photos or video streams. The machine learning capability is the essential ingredient of artificial intelligence but general artificial intelligence assumes a  full range of human cognitive capabilities.

The applications of machine learning are broadly used today in many areas.

The current state of Machine Learning

According to the recent survey conducted by MIT Technology Review Custom and Google Cloud (2):

  • ML is happening now. The majority of respondents (60 percent) have already implemented ML strategies, and nearly one-third considered themselves to be at a mature stage with their initiatives.
  • ML provides marketplace advantage. According to respondents, a key benefit of ML is the ability to gain a competitive edge, and 26 percent of current ML implementers felt they had already achieved that goal.
  • Organizations are investing in ML. Among current ML implementers, some 26 percent reported that more than 15 percent of their IT budgets was dedicated to ML initiatives.
  • Early adopters are realizing ML’s biggest potential benefits. The top hoped for benefit among ML implementers and planners is the ability to extend data analysis efforts and increase data insights. Some 45 percent of respondents report success in meeting that goal. In addition, more than half of both early-stage and mature-stage users say their ML efforts have resulted in demonstrable return on investment (ROI).
  • ML implementers are pursuing a broad range of projects. The most common projects among current ML implementers are image recognition, classification, and tagging (47 percent); emotion/behavior analysis (47 percent); text classification and mining (47 percent); and natural language processing, or NLP (45 percent).

Applying ML technology today

Below are the areas just examples of the business operations areas where the retailers main gain significant benefits from implementing ML technology today.

  • Demand prediction and proactive inventory management dramatically reducing inventory manage cost
  • Personalization and recommendation engine improves online customer experience and help generate more revenue
  • Pricing optimization algorithms contribute to staying competitive and profitable at the same time.
  • Robotics helps in warehouse automation and timely orders fulfillment.
  • Machine learning fraud prevention helps eCommerce operators eliminate or reduce losses due to online fraud.
  • Cybersecurity