ramekin | An open source , real time trend detection library | Machine Learning library

ramekin is a JavaScript library typically used in Artificial Intelligence, Machine Learning, Deep Learning, Tensorflow applications. ramekin has no bugs, it has no vulnerabilities, it has a Permissive License and it has low support. You can install using 'npm i ramekin' or download it from GitHub, npm.

An open source, real time trend detection framework. This project uses machine learning to detect trends in text over time. Trends are identified by detecting phrases that start occurring much more frequently than those that don't typically occur. Various natural language processing and data science techniques are used to ensure similar words are modelled together (i.e. "cycle", "cycling" and "cyclist" all reduce down to a common word form, such as "cycle"). Documents can be grouped by a subject, so it is possible to detect "localised" trends - for example, 7 articles talking about a new bike from Santa Cruz might. Similar phrases tend to relate to a particular trend, so hierachical clustering is used to make sure documents related to the same trend are grouped, rather than creating two "trends" about the same thing. For example, "doping scandal" and "Tour de France" are likely to be about the same thing...allegedly. Keywords: trending, trends, news, natural language processing, NLP, machine learning, artificial intelligence, data science, hierarchical clustering.