HTM for Stocks
Find anomalies in publicly traded stocks using trading and Twitter data.
HTM for Stocks is an example HTM application that continually monitors hundreds of publicly traded companies and alerts you if something unusual is happening to any of them. HTM for Stocks uses HTM machine intelligence algorithms to model stock price, stock volume, and Twitter data related to 200 of the largest publicly traded companies. Companies monitored include Apple, Google, Amazon, and Starbucks. HTM for Stocks is a mobile application that runs on both iOS and Android-based phones.
Numenta has made the source code for HTM for Stocks available alongside the NuPIC open source project to encourage others to create new and derivative products. Contact us with questions or interest. Review the source code.
Time-based Pattern Detection
Automatically analyzes time-based patterns on a company by company basis to identify anomalies. The HTM algorithm ensures that reported anomalies are rare (avoiding false positives), and that real anomalies can be detected, even if humans can’t see them (avoiding false negatives).
Most Anomalous Companies Highlighted
Companies are listed in order of most anomalous in the past 24 hours so everything the user wants to see is usually visible on one page.
Learns with each data point and adapts to changes over time.
Displays tweets around the same time as anomalies so you can easily see what people are saying and why something unusual is happening.
Sends alerts when anomalies occur on your favorite companies.
iOS and Android Mobile App
Presents output in a friendly, mobile user interface.
How Does It Work?
Ingests real-time stock and Twitter data
For each monitored company, data is collected every five minutes for stock price, stock volume and Twitter volume. Actual data charts can be viewed in the application.
Models build automatically
Models are automatically built and refined for each data stream. No data scientists are required!
Anomalies are based on deviation from recent performance
Each company is different. HTM models report anomalies when current behavior deviates from recent behavior.
Sample Use Cases
CEO wants to be alerted about unusual activity with competitors.
Sales Rep monitors companies in her territory to harvest new information for sales calls.
Marketing Executive aims to understand product announcement chatter.
Financial market followers track companies of interest, noting when anomalies occur.