mi_filter_feat

Performing feature selection in a automatic way based on correlation and feature importance.

The number of biological databases is growing rapidly, but different databases use different IDs to refer to the same biological entity. The inconsistency in IDs impedes the integration of various types of biological data. To resolve the problem, we developed 'MantaID', a data-driven, machine-learning based approach that automates identifying IDs on a large scale. The 'MantaID' model's prediction accuracy was proven to be 99%, and it correctly and effectively predicted 100,000 ID entries within two minutes. 'MantaID' supports the discovery and exploitation of ID patterns from large quantities of databases. (e.g., up to 542 biological databases). An easy-to-use freely available open-source software R package, a user-friendly web application, and API were also developed for 'MantaID' to improve applicability. To our knowledge, 'MantaID' is the first tool that enables an automatic, quick, accurate, and comprehensive identification of large quantities of IDs, and can therefore be used as a starting point to facilitate the complex assimilation and aggregation of biological data across diverse databases.

Zeng Zhengpeng

MantaID

A Machine-Learning Based Tool to Automate the Identification of
Biological Database IDs

Zhengpeng Zeng

Longfei Mao

Feng Yu

Jiamin Hu

Xiting Wang

mi_filter_feat function

<dl><dt>data</dt>
<dd>The data frame returned by <code>mi_to_numer()</code>.</dd>
<dt>cor_thresh</dt>
<dd>The threshold set for Pearson correlation. If correlation value is over this threshold, the two features will be viewed as redundant and one of them will be removed.</dd>
<dt>imp_thresh</dt>
<dd>The threshold set for feature importance. The last several features with the lowest importance will be removed if remained importance lower than <code>imp_thresh</code>.</dd>
<dt>union</dt>
<dd>The method for combining the decisions of correlation method and importance method. If <code>TRUE</code>, any of the features calculated by the two methods will be returned. Otherwise, only features in the results of both methods will be returned.</dd></dl>

Arguments

Performing feature selection in a automatic way based on correlation and feature importance. — mi_filter_feat

<dl>

<dt>data</dt>
<dd>The data frame returned by <code>mi_to_numer()</code>.</dd>


<dt>cor_thresh</dt>
<dd>The threshold set for Pearson correlation. If correlation value is over this threshold, the two features will be viewed as redundant and one of them will be removed.</dd>


<dt>imp_thresh</dt>
<dd>The threshold set for feature importance. The last several features with the lowest importance will be removed if remained importance lower than <code>imp_thresh</code>.</dd>


<dt>union</dt>
<dd>The method for combining the decisions of correlation method and importance method. If <code>TRUE</code>, any of the features calculated by the two methods will be returned. Otherwise, only features in the results of both methods will be returned.</dd>

</dl>

mi_filter_feat: Performing feature selection in a automatic way based on correlation and feature importance.

Description

Usage

Value

Arguments