<?xml version="1.0" encoding="UTF-8"?>
<ArticleSet>
  <Article>
    <Journal>
      <PublisherName>KMAN Publication Inc. (KMANPUB)</PublisherName>
      <JournalTitle>Journal of Assessment and Research in Applied Counseling (JARAC)</JournalTitle>
      <Issn></Issn>
      <Volume>8</Volume>
      <Issue>Serial Number 29</Issue>
      <PubDate PubStatus="epublish">
        <Year>2026</Year>
        <Month>04</Month>
        <Day>01</Day>
      </PubDate>
    </Journal>
    <ArticleTitle>Machine‑Learning Prediction of Academic Stress from Working Memory Load and Neuroticism Facets</ArticleTitle>
    <VernacularTitle>Machine‑Learning Prediction of Academic Stress from Working Memory Load and Neuroticism Facets</VernacularTitle>
    <FirstPage>1</FirstPage>
    <LastPage>11</LastPage>
    <ELocationID EIdType="doi">10.61838/kman.jarac.5166</ELocationID>
    <Language>EN</Language>
    <AuthorList>
      <Author>
        <FirstName></FirstName>
        <LastName></LastName>
        <Affiliation></Affiliation>
      </Author>
      <Author>
        <FirstName></FirstName>
        <LastName></LastName>
        <Affiliation></Affiliation>
      </Author>
    </AuthorList>
    <PublicationType>Journal Article</PublicationType>
    <History>
      <PubDate PubStatus="received">
        <Year>2025</Year>
        <Month>12</Month>
        <Day>23</Day>
      </PubDate>
    </History>
    <Abstract>&lt;p&gt;&lt;strong&gt;Objective:&lt;/strong&gt; To employ advanced machine learning algorithms to predict the severity of academic stress among university students by synthesizing metrics of working memory load and specific facets of the neuroticism personality trait.&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Methods and Materials: &lt;/strong&gt;This predictive cross-sectional study evaluated a sample of N=452 Brazilian university students. Data were collected utilizing the Perception of Academic Stress Scale, a dual N-back task to measure working memory capacity, and the NEO Personality Inventory-Revised to assess neuroticism facets. The predictive modeling utilized several machine learning algorithms, with a primary focus on eXtreme Gradient Boosting (XGBoost). The dataset was split into an 80%training set with 10-fold cross-validation and a 20%hold-out test set, and feature importance was interpreted using SHapley Additive exPlanations (SHAP) values.&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Findings: &lt;/strong&gt;The XGBoost algorithm yielded the most robust predictive performance, explaining a significant portion of the variance in the test set (R^2=0.58, MAE=4.12, RMSE=5.36). SHAP analysis revealed that the Vulnerability to Stress facet was the strongest positive predictor of academic stress (mean absolute SHAP = 2.84), whereas Working Memory Capacity emerged as the strongest negative, protective predictor (mean absolute SHAP = 1.95), specifically when surpassing a standardized threshold of Z=0.50.&lt;/p&gt;
&lt;p&gt;&lt;strong&gt;Conclusion: &lt;/strong&gt;Machine learning architectures successfully demonstrate that academic stress is deeply driven by non-linear interactions between a student’s intrinsic emotional vulnerability and their momentary cognitive working memory capacity.&lt;/p&gt;</Abstract>
    <ObjectList>
      <Object Type="keyword">
        <Param Name="value">Academic Stress</Param>
      </Object>
      <Object Type="keyword">
        <Param Name="value">Working Memory</Param>
      </Object>
      <Object Type="keyword">
        <Param Name="value">Neuroticism</Param>
      </Object>
      <Object Type="keyword">
        <Param Name="value">Machine Learning</Param>
      </Object>
      <Object Type="keyword">
        <Param Name="value">Students</Param>
      </Object>
    </ObjectList>
    <ArchiveCopySource DocType="pdf">https://journals.kmanpub.com/index.php/jarac/article/download/5166/9453</ArchiveCopySource>
  </Article>
</ArticleSet>
