AI Methods (and Human Review). LearnerShape uses artificial intelligence (AI) based on a combination of machine learning and other data science methods to recommend courses that teach specific skills. We are steadily advancing our AI methods. Current methods including the BERT algorithm (developed by Google), logistic regression and random forest classification. We also currently use a final human review of recommendations (we expect to reduce the need for human review over time).
Recommending Courses for Any Skill. Our use of AI allows us to efficiently recommend any course content for any skill, so long as both the course and skill have a textual description (which is almost always the case). The ability to use any skill efficiently means that individual organizations can use their own skills frameworks (which we call 'skills graphs' on our platform), massively increasing the effectiveness of LearnerShape over existing solutions. Please contact us to learn more.
Ensuring Course Quality. We aim to ensure course quality through various methods, including:
reliable AI methods
reliable content providers
use of data from content providers and our own customers on course quality
human review of AI-based recommendations.
We are also working on additional AI-based techniques for assessing course quality, as well as course level (so that we recommend appropriate courses based on your current and target skill levels).
Problems with Recommendations. We know that we won't always get recommendations exactly right. Recommendation of learning content is a challenging task with inherent subjectivity, either for a human or an AI. As we improve our AI methods, we want to know when there is a problem with our recommendations. The best way to tell us about a problem is to use the 'Report a problem' button on the course information pane. You can also contact LearnerShape by any other method to provide information or comments on course quality.