fast_find_neighbors

This function finds the k nearest neighbors for each row in a matrix using the specified distance metric. Distance metrics available are "p-norm", "Chebyshev", "Canberra", "Overlap", "HEOM" (Heterogeneous Euclidean-Overlap Metric), "HVDM" (Heterogeneous Value Difference Metric). See the <code>calculate_distance</code> function of our package for more details on the distances.

Functions to classify mass spectra in known categories, and to determine discriminant mass-to-charge values. It includes easy-to-use functions for preprocessing mass spectra, functions to determine discriminant mass-to-charge values (m/z) from a library of mass spectra corresponding to different categories, and functions to predict the category (species, phenotypes, etc.) associated to a mass spectrum from a list of selected mass-to-charge values. If you use this package in your research, please cite the associated publication (<doi:10.1016/j.eswa.2025.128796>). For a comprehensive guide, additional applications, and detailed examples of using this package, please visit our GitHub repository (<https://github.com/agodmer/MSclassifR_examples>).

Alexandre Godmer

MSclassifR

Automated Classification of Mass Spectra

Quentin Giai Gianetto

Karen Druart

fast_find_neighbors function

<dl> <dt>data</dt>
<dd>Matrix where the k nearest neighbors for each row are searched.</dd>
 <dt>nominal_indices</dt>
<dd>Vector of column indices indicating which features are
 categorical (nominal) variables. This is crucial for proper distance calculation
 as nominal and numeric features require different handling. For example, if columns
 2 and 5 contain categorical variables, nominal_indices should be <code>c(2, 5)</code>. See <code>calculate_distance</code> function.</dd>
 <dt>p_code</dt>
<dd>Numeric code representing the distance metric to use:<ul>
<li>p &gt;= 1: p-norm</li>
<li>p = 0: Chebyshev</li>
<li>p = -1: Canberra</li>
<li>p = -2: Overlap (nominal attributes only)</li>
<li>p = -3: HEOM (Heterogeneous Euclidean-Overlap Metric)</li>
<li>p = -4: HVDM (Heterogeneous Value Difference Metric)</li>
</ul></dd>
 <dt>k</dt>
<dd>Number of nearest neighbors to find.</dd>
</dl>

Arguments

Function finding k Nearest Neighbors for each row of a matrix — fast_find_neighbors

<dl>

 <dt>data</dt>
<dd>Matrix where the k nearest neighbors for each row are searched.</dd>


 <dt>nominal_indices</dt>
<dd>Vector of column indices indicating which features are
 categorical (nominal) variables. This is crucial for proper distance calculation
 as nominal and numeric features require different handling. For example, if columns
 2 and 5 contain categorical variables, nominal_indices should be <code>c(2, 5)</code>. See <code>calculate_distance</code> function.</dd>


 <dt>p_code</dt>
<dd>Numeric code representing the distance metric to use:<ul>
<li>p &gt;= 1: p-norm</li>
<li>p = 0: Chebyshev</li>
<li>p = -1: Canberra</li>
<li>p = -2: Overlap (nominal attributes only)</li>
<li>p = -3: HEOM (Heterogeneous Euclidean-Overlap Metric)</li>
<li>p = -4: HVDM (Heterogeneous Value Difference Metric)</li>
</ul></dd>


 <dt>k</dt>
<dd>Number of nearest neighbors to find.</dd>


</dl>

fast_find_neighbors: Function finding k Nearest Neighbors for each row of a matrix

Description

Usage

Value

Arguments

See Also