Mona, the maker of an AI monitoring platform, has introduced a new automated exploratory data analysis tool to identify the root cause of anomalies in multivariate datasets.
Multivariate data analysis can be complex and time consuming, especially when drawing insights from large datasets, and Mona says its new exploratory tool is designed to streamline and simplify this process. The tool uses an algorithm to automate the process of finding specific segments in large datasets where metrics underperform or exhibit anomalous behaviors while avoiding noise and false positives. The tool then allows for further analysis of the possible reasons: “By correlating key findings with other relevant metrics in the dataset, Mona is able to provide a comprehensive understanding of underlying patterns to determine the true cause of an anomaly,” the company said in a statement.
“In many industries, a significant part of analysts’ work is to find the specific segments in which metrics underperform, and then understand the reason behind it,” said Itai Bar-Sinai, CPO & co-founder at Mona. “We created the first-ever algorithm to automate this process on any multivariate dataset.”
Mona has offices in Tel Aviv and Atlanta and was founded in 2018 by Google and McKinsey & Co. alums Yotam Oren, Nimrod Tamir, and Itai Bar Sinai. “About 3 years ago, we ventured on a mission to empower data teams to make AI more impactful and reliable, and to raise the collective confidence of business and technology leaders in their ability to make the most out of AI,” says Mona’s website.
The company specializes in AI monitoring and automation tools which it says provide data and AI teams with continuous insights to help reduce risks, optimize operations, and build more valuable AI systems. Mona recently introduced a new feature of its monitoring platform used for assessing AI fairness. The feature provides a comprehensive view of an AI system and automatically detects potential issues of bias, according to the company. Users can then generate fairness reports that include standard and custom options and are applicable for internal and external auditing.
Mona currently has a sign-up for beta access to its new exploratory data analysis tool at this link, as well as a demonstration video.
“As organizations increasingly rely on data to inform their decision making, they need more efficient and accurate data analysis tools,” said Oren. “We’re excited to make Mona’s cutting-edge analytical capabilities more accessible to more data practitioners.”
Observability Primed for a Breakout 2023: Prediction
Solving for Data Drift from Class Imbalance with Model Monitoring
Your Data Is Talking. Are You Listening?