[Sometimes called Case-Based Reasoning or CBR]
"At the highest level of generality, a general CBR cycle may be described by the following four processes: 1. RETRIEVE the most similar case or cases. 2. REUSE the information and knowledge in that case to solve the problem. 3. REVISE the proposed solution. 4. RETAIN the parts of this experience likely to be useful for future problem solving "– from Case-Based Reasoning: Foundational Issues, Methodological Variations, and System Approaches. By A. Aamodt and E. Plaza. (1994)
Intuit recently announced QuickBooks Capital, a new small business lending product that provides users of QuickBooks access to small business loans up to about $35,000. This new service's lending process is done algorithmically and from within QuickBooks itself. Thanks to big data and machine learning techniques, most borrowers will know whether or not they are approved for a loan in just a a few minutes. Lack of credit is consistently one of the top challenges for small businesses. And for new businesses (those that have been in business less than 5 years), the challenges are even greater.
There is no stronger marketing tool than customer and member insight. The key is applying this insight to move beyond great internal reports to enhance marketing and the consumer experience on a daily basis (or in real time). Two mothers sit on a park bench, watching their offspring on the playground. Mom No. 1 observes to Mom No. 2, "Your son is very intelligent." Mom No. 2 sighs, shakes her head and responds, "Yeah, but he's not interested in learning."
In other words, the models are trained to recognize good software and bad software in order to block the bad. This causes training models to recognize the differences between malware and common packaged software, yet ignore the profile of custom or lesser-known applications that may also be present. According to a Barkly survey of IT administrators, 42 percent of companies believe that their users lost productivity as a result of false positive results. To provide maximum value while reducing the pressure on overworked staff, security based on machine learning must balance blocking malicious software with avoiding impact on the regular use of business applications.
"The system could potentially be an aid for doctors in the ICU, which is a high-stress, high-demand environment," says PhD student Harini Suresh, lead author on the paper about ICU Intervene. "The goal is to leverage data from medical records to improve health care and predict actionable interventions." Another team developed an approach called "EHR Model Transfer" that can facilitate the application of predictive models on an electronic health record (EHR) system, despite being trained on data from a different EHR system. "Much of the previous work in clinical decision-making has focused on outcomes such as mortality (likelihood of death), while this work predicts actionable treatments," Suresh says.
"The goal is to leverage data from medical records to improve health care and predict actionable interventions." Another team developed an approach called "EHR Model Transfer" that can facilitate the application of predictive models on an electronic health record (EHR) system, despite being trained on data from a different EHR system. EHR Model Transfer was found to outperform baseline approaches and demonstrated better transfer of predictive models across EHR versions compared to using EHR-specific events alone. In the future, the EHR Model Transfer team plans to evaluate the system on data and EHR systems from other hospitals and care settings.
As defensive technologies based on machine learning become increasingly numerous, so will offensive ones – whether wielded by attackers or pentesters. We guarantee that you'll be either writing machine learning hacking tools next year, or desperately attempting to defend against them," the researchers concluded. At the same conference, Hyrum Anderson, Technical Director of Data Science at Endgame, explained how an AI agent trained through reinforcement learning to modify malware can successfully evade machine learning malware detection. As DeepHack, Anderson's AI agent was able to "learn" by playing thousands of "games."
As part of the update to Google Forms, one of the improvements included is intelligent response validation, and from time to time (whenever it's possible to do so) Google Forms will make a suggestion to users to validate a response that was issued by the person filling out a Google Form based on the questions that are asked by the form's creator. Also in the presence of saving time for users, Google Forms will now allow you to set up pre-configured preferences for future forms that you create so you don't have to choose certain elements each time you set up a new form, such as the option for always collecting email addresses or making questions required. Google has set limits on the file uploads, which starts at just 1GB, but there's also an option to increase the limit to 1TB if it's needed. So, when creating a new Form, if you want to provide the recipients with the ability to select multiple options for a single question, the Checkbox Grid would be the one to pick.
In threat trapping, passive technologies identify malware using models of bad behavior like signatures. Unfortunately, developing accurate malware detection products based on good behavior modeling is not easy. But no company has enough human resources to manually evaluate a large number of alerts about possible security threats. When AI applies both bad and good behavior models, it reduces the number of false positives to a manageable amount.
Campus Technology reports that a team of educators developed a writing-to-learn tool called M-Write, which uses automated text analysis (ATA) to identify the strengths of a writing submission. A report from the EDUCAUSE Center for Analysis and Research explores how the University of Central Florida piloted adaptive learning in large, introductory courses like General Psychology. General Psychology is a general education course with many sections that can often be taught by adjuncts. "As with any new tool, adaptive learning provides a new set of capabilities and insights -- and a lot of very useful data -- that can be used to explore ways to increase students learning and success," ECAR reports.