Z-Scores and Data Classification

Summary of my bookmarked Github repositories from Feb 16th, 2017

Github repositories

  • kcrt/stat-lms

    stat-lms is an npm package that allows you to calculate z-scores using the LMS (lambda-mu-sigma) method. It is particularly useful for skewed distributions where the data does not follow a normal distribution. The package provides functions such as `getZScore` to calculate the z-score for a given value, `getPercentile` to determine the percentile corresponding to a z-score, and `getValueFromZScore` to calculate the value corresponding to a given z-score. It was programmed by TAKAHASHI, Kyohei and is released under the MIT license.

  • NathanEpstein/KNear

    kNear is a JavaScript implementation of the k-nearest neighbors algorithm, a supervised machine learning algorithm. It assigns classifications to numeric points based on learned data. By calculating the proximity of new points to the learned points, it can classify them accordingly. The usage involves setting the value of k (an integer), learning data points with their respective classifications, and then using the machine to classify new points. The library is available on npm and is accompanied by the MIT License.

  • Zeikko/z-score

    The Z-Score library is a tool that calculates the z-score for numeric object attributes. With a simple installation process (npm install z-score), you can utilize this library to obtain z-scores for your data. By training the library with a set of objects containing attribute values, you can then calculate the z-score for a specific object. The example demonstrates the calculation of the z-score for a temperature of 15 and a pressure of 1122, resulting in a z-score of -2 for temperature and -3 for pressure.