The fields of machine learning and deep learning are on the brink of unprecedented breakthroughs across a variety of verticals. And according to a new report from Databricks, “data is the new fuel,” for these market advancements.
AI and deep learning are set to disrupt and change industries across the board, and potential for innovation is certainly great. That said, the question for many enterprises becomes how to take advantage of the myriad of data ML tools now available.
The new report explores the Databricks Unified Analytics Platform and provides four real-life machine learning use cases, including code samples and notebooks.
The platform is a cloud-service designed to provide users with “ready-to-use clusters that can handle all analytics processes in one place,” the report explains. These processes range from data preparation to model building and serving.
The new guide walks readers through four practical end-to-end machine learning use cases on Databricks.
The first scenario explored is a loan risk analysis use case that covers importing and exploring data in Databricks. The white paper points out that being able to assess the risk of loan applications can save a lender the cost of holding too many risky assets.
Next up, the report explores an advertising analytics and click prediction use case, including collecting and exploring the advertising logs with Spark SQL, using PySpark for feature engineering and using GBTClassifier for model training and predicting the clicks.
Advanced analytics, including but not limited to classification, clustering, recognition, prediction and recommendations allow ad and analytics companies to gain deeper insights from their data and drive business outcomes.
The third use case involves a market basket analysis — a key technique to uncover associations between different items — problem at scale.
“With the rapid growth e-commerce data, it is necessary to execute models like market basket analysis on increasing larger sizes of data,” the report states.
Lastly, Databricks breaks down suspicious behavior identification in a video example, including pre-processing step to create image frames, transfer learning for featurization, and applying logistic regression to identify suspicious images in a video. It is becoming increasingly important today to operationalize and automate the process of video identification and categorization.
Download the new white paper today, “Four Real-Life Machine Learning Use Cases,” to explore Databricks Unified Analytics Platform use cases in the advertising, loan servicing, media industries and more.