A list of cybersecurity projects that use machine learning techniques
| Company | Description | Methods and technology |
|---|---|---|
| Cylance (https://www.cylance.com/) | Cylance is revolutionizing cybersecurity with products and services that proactively prevent, rather than reactively detect the execution of advanced persistent threats and malware. CylancePROTECT’s architecture consists of a small agent that integrates with existing software management systems or Cylance’s own cloud console. The endpoint will detect and prevent malware through the use of tested mathematical models on the host, independent of a cloud or signatures. It is capable of detecting and quarantining malware in both open and isolated networks without the need for continual signature updates. Defense requires applying the best protection at the most vulnerable locations - the endpoints. Cylance’s mathematical approach stops the execution of harmful code regardless of having prior knowledge or employing an unknown obfuscation technique. No other anti-malware product compares to the accuracy, ease of management and effectiveness of CylancePROTECT. | |
| Darktrace (https://www.darktrace.com/) | Using new machine learning techniques based on the biological principles of the human immune system, Darktrace addresses the challenge of detecting previously unidentified cyber threats, irrespective of their origin.Immune computing. Darktrace’s unique technology is powered by advanced machine learning, allowing it to learn what is normal for a company’s network environment, so that it can then determine if any behavior is abnormal. This allows it to detect cyber attacks of a nature that may not have been observed before, the unknown unknowns. The ability to self-learn and adapt to a changing environment in real-time allows organizations to reconcile the need for an interconnected workforce, customer base and supply chain, whilst ensuring that they protect against serious, existential threats to their businesses in the most effective and pragmatic way possible. | Bayes |
| Zimperium (https://www.zimperium.com/) | Zimperium developed the innovative and award-winning z9 engine by training it over many years on proprietary machine-learning algorithms that distinguish normal from malicious behavior on Android and iOS devices. Unlike cloud-based mobile security solutions that employ app sandboxing or tunnel traffic through the cloud, the z9 engine sits directly on the mobile devices within the zIPS app, in order to provide complete protection around the clock. | |
| Deepinstinct (www.deepinstinct.com/) | Anti-threat and Antivirus | |
| Sentinel (https://sentinelone.com/) | Paper - https://sentinelone.com/wp-content/uploads/2016/07/SentinelOne-EPP-Data-Sheet-FINAL-JUNE-2016.pdf | |
| Demisto (https://www.demisto.com/) | An open community for digital forensics and incident response security analysts to share tools, process and playbooks. | |
| Graphistry (https://www.graphistry.com/) | Visually analyze millions of events on a graph.Data, Splunk, REST API. | Using NVIDIA GPUs and CUDA, the graph analysis cloud platform is able to help the company’s response and hunting team sift through 100M+ alerts a day.https://www.youtube.com/watch?v=ZA9Ityyk4pg |
| SignalSence (www.signalsense.com/) | Paper - http://www.ten-inc.com/presentations/deep_learning.pdf | |
| Appzen (https://www.appzen.com/) | AppZen, the first artificial intelligence (AI) solution for back office automation, First Automated Expense Report Auditing and Compliance Solution. With AppZen, which uses patent-pending Natural Language Processing (NLP) machine learning algorithms, companies can review expense report data, cross-check it with external sources and social media in real-time, and detect compliance problems before it is too late. The automated tool, which also offers a “virtual assistant” function that helps employees gather expenses for reporting, lowers expense processing costs up to 80 percent and identifies up to 10 times more financial risks. |