<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.2 20190208//EN" "JATS-archivearticle1.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:mml="http://www.w3.org/1998/Math/MathML" article-type="research-article" dtd-version="1.2" xml:lang="az">
  <front>
    <journal-meta>
      <journal-id journal-id-type="publisher-id">ndujournal</journal-id>
      <journal-title-group>
        <journal-title>NDU Journal of Natural Resources</journal-title>
      </journal-title-group>
      <issn pub-type="epub">2225-2401</issn>
      <publisher>
        <publisher-name>Naxcivan Dovlet Universiteti</publisher-name>
      </publisher>
    </journal-meta>
    <article-meta>
      <article-id pub-id-type="publisher-id">9</article-id>
            <article-id pub-id-type="doi">10.3390/app15073792414</article-id>
            <title-group>
        <article-title xml:lang="az">IOT ƏSASLI AĞILLI KİTABXANA MÜHİTLƏRİNDƏ SENSOR ƏHƏMİYYƏTİNİN İNTERPRETASİYASI VƏ MAŞIN ÖYRƏNMƏSİ MODELLƏRİNİN HESABLAMA SƏMƏRƏLİLİYİNİN MÜQAYİSƏLİ TƏHLİLİ</article-title>
                <article-title xml:lang="en">INTERPRETATİON OF SENSOR IMPORTANCE AND COMPARATİVE ANALYSİS OF COMPUTATİONAL EFFİCİENCY OF MACHİNE LEARNİNG MODELS İN IOT-BASED SMART LİBRARY ENVİRONMENTS</article-title>
              </title-group>
      <contrib-group>
                <contrib contrib-type="author">
          <name>
            <surname>Sərkan</surname>
            <given-names>Məmmədov</given-names>
          </name>
                    <email>serkan.m@ndu.edu.az</email>
                              <contrib-id contrib-id-type="orcid">https://orcid.org/0009-0007-9413-0971</contrib-id>
                              <aff>Naxçıvan Dövlət Universiteti</aff>
                  </contrib>
              </contrib-group>
      <pub-date pub-type="pub">
        <year>2026</year>
        <month>06</month>
        <day>24</day>
      </pub-date>
            <volume>1</volume>
      <issue>1</issue>
            <fpage>1</fpage>
      <lpage>12</lpage>
      <permissions>
        <license license-type="open-access" xlink:href="http://creativecommons.org/licenses/by/4.0/">
          <license-p>This is an open-access article distributed under the terms of the Creative Commons Attribution License.</license-p>
        </license>
      </permissions>
      <abstract xml:lang="az">
        <p>Bu tədqiqat IoT əsaslı ağıllı kitabxana mühitlərində ətraf mühit sensorlarının maşın öyrənməsi modellərinin qərarvermə prosesinə təsirinin interpretasiyasını və hesablama səmərəliliyinin qiymətləndirilməsini hədəfləyir. Düzce Universiteti Kitabxanasında toplanmış çoxölçülü sensor məlumatları əsasında yeddi fərqli maşın öyrənməsi alqoritmi KNN, Random Forest, Decision Tree, SVM, XGBoost, Logistic Regression və Naive Bayes  müqayisəli şəkildə tətbiq edilmişdir. Tədqiqatın əsas elmi töhfəsi sensor əhəmiyyətinin qiymətləndirilməsində iki yanaşmanın, yəni Random Forest Feature Importance və Permutation Importance metodlarının paralel istifadəsidir. Bu yanaşma sensor dəyişənlərinin model daxili qərar mexanizmi ilə real performansa təsirini daha dəqiq analiz etməyə imkan vermişdir. Nəticələr göstərir ki, Feature Importance yanaşması temperatur və insan sıxlığı kimi dəyişənləri daha dominant faktorlar kimi qiymətləndirərkən, Permutation Importance analizi işıq intensivliyi və CO₂ əsaslı dəyişənlərin model performansına daha yüksək təsir göstərdiyini ortaya qoyur. Bu fərqlilik sensor interpretasiyasında yalnız model daxili ölçülərin kifayət etmədiyini və daha etibarlı qiymətləndirmə metodlarına ehtiyac olduğunu göstərir. Bundan əlavə, hesablama səmərəliliyi baxımından KNN və Decision Tree modelləri aşağı resurs tələbi və yüksək icra sürəti ilə real vaxt IoT tətbiqləri üçün daha uyğun nəticələr nümayiş etdirmişdir. Ümumilikdə tədqiqat göstərir ki, IoT əsaslı sistemlərdə optimal model seçimi yalnız proqnoz dəqiqliyi deyil, həm də sensor interpretasiyası və hesablama səmərəliliyi birlikdə nəzərə alınaraq aparılmalıdır.</p>
      </abstract>
            <abstract xml:lang="en">
        <p>This study aims to interpret the impact of environmental sensors on the decision-making processes of machine learning models in IoT-based smart library environments and to evaluate computational efficiency. Based on multidimensional sensor data collected from Düzce University Library, seven machine learning algorithms — KNN, Random Forest, Decision Tree, SVM, XGBoost, Logistic Regression, and Naïve Bayes — were comparatively applied. The primary scientific contribution of this study is the parallel implementation of two sensor importance assessment approaches, namely Random Forest Feature Importance and Permutation Importance. This dual-method framework enabled a more precise analysis of the relationship between model-internal decision mechanisms and the actual impact of sensor variables on predictive performance. The results indicate that while the Feature Importance approach identifies temperature and human occupancy as dominant factors, Permutation Importance analysis reveals that light intensity and CO₂-based variables exert a greater influence on model performance. This discrepancy demonstrates that relying solely on model-internal importance measures is insufficient for accurate sensor interpretation, highlighting the need for more robust evaluation methodologies. Furthermore, in terms of computational efficiency, KNN and Decision Tree models exhibited more suitable characteristics for real-time IoT applications, owing to their low resource consumption and high execution speed. Overall, the findings suggest that optimal model selection in IoT-based systems should be guided not only by predictive accuracy, but also by sensor interpretability and computational efficiency considered in an integrated manner.</p>
      </abstract>
            <kwd-group xml:lang="az">
                              <kwd>IoT</kwd>
                      <kwd>Maşın öyrənməsi</kwd>
                      <kwd>Sensor analizi</kwd>
                      <kwd>Feature importance</kwd>
                      <kwd>Permutation importance</kwd>
                        </kwd-group>
      <kwd-group xml:lang="en">
                              <kwd>IoT</kwd>
                      <kwd>Machine Learning</kwd>
                      <kwd>Sensor Analysis</kwd>
                      <kwd>Feature Importance</kwd>
                      <kwd>Permutation Importance</kwd>
                        </kwd-group>
    </article-meta>
  </front>
  <body>
    <p>Məqalə mətni PDF formatında yüklənə bilər.</p>
  </body>
  <back>
    <ref-list>
            <ref id="ref1">
        <element-citation publication-type="journal">
          <text>Akanmu, W. P., Nunayon, S. S., &amp; Eboson, U. C. (2021). Indoor environmental quality (IEQ) assessment of Nigerian university libraries: A pilot study. Energy and Built Environment, 2(3), 302–314. https://doi.org/10.1016/J.ENBENV.2020.07.004</text>
                    <pub-id pub-id-type="doi">10.1016/J.ENBENV.2020.07.004</pub-id>
                  </element-citation>
      </ref>
            <ref id="ref2">
        <element-citation publication-type="journal">
          <text>Bischo, W., &amp; Lahrz, T. (2008). Gesundheitliche Bewertung von Kohlendioxid in der Innenraumluft. Bundesgesundheitsblatt - Gesundheitsforschung - Gesundheitsschutz 2008 51:11, 51(11), 1358–1369. https://doi.org/10.1007/S00103-008-0707-2</text>
                    <pub-id pub-id-type="doi">10.1007/S00103-008-0707-2</pub-id>
                  </element-citation>
      </ref>
            <ref id="ref3">
        <element-citation publication-type="journal">
          <text>Elreedy, D., Amir, ·, Atiya, F., Kamalov, · Firuz, Moniz, N., Branco, P., Torgo, L., Japkowicz, N., Wozniak, M., Wang, S., Atiya, A. F., &amp; Kamalov, F. (2023). A theoretical distribution analysis of synthetic minority oversampling technique (SMOTE) for imbalanced learning. Machine Learning 2023 113:7, 113(7), 4903–4923. https://doi.org/10.1007/S10994-022-06296-4</text>
                    <pub-id pub-id-type="doi">10.1007/S10994-022-06296-4</pub-id>
                  </element-citation>
      </ref>
            <ref id="ref4">
        <element-citation publication-type="journal">
          <text>Hong, S., Kim, Y., &amp; Yang, E. (2022). Indoor environment and student productivity for individual and collaborative work in learning commons: a case study. Library Management, 43(1–2), 15–34. https://doi.org/10.1108/LM-06-2021-0055</text>
                    <pub-id pub-id-type="doi">10.1108/LM-06-2021-0055</pub-id>
                  </element-citation>
      </ref>
            <ref id="ref5">
        <element-citation publication-type="journal">
          <text>Hoşten, G., &amp; Dalbay, N. (2018). Kapalı Ortam Hava Kalitesinin Ofis Ortamlarında İş Sağlığı ve Güvenliği Açısından Değerlendirilmesi. Aydın Journal of Health, 4(2), 1–12. https://dergipark.org.tr/en/pub/asder/article/530250</text>
                  </element-citation>
      </ref>
            <ref id="ref6">
        <element-citation publication-type="journal">
          <text>Khan, A. U., Zhang, Z., Chohan, S. R., &amp; Rafique, W. (2022). Factors fostering the success of IoT services in academic libraries: a study built to enhance the library performance. Library Hi Tech, 40(6), 1976–1995. https://doi.org/10.1108/LHT-06-2021-0179</text>
                    <pub-id pub-id-type="doi">10.1108/LHT-06-2021-0179</pub-id>
                  </element-citation>
      </ref>
            <ref id="ref7">
        <element-citation publication-type="journal">
          <text>Khritish, S. (2023). The Impact of Study Environment on Students’ Academic Performance: An Experimental Research Study. https://doi.org/https://doi.org/10.36227/techrxiv.170250911.12566888/v1</text>
                    <pub-id pub-id-type="doi">10.36227/techrxiv.170250911.12566888/v1</pub-id>
                  </element-citation>
      </ref>
            <ref id="ref8">
        <element-citation publication-type="journal">
          <text>Maashi, M., Alabdulkreem, E., Maray, M., Shankar, K., Darem, A. A., Alzahrani, A., &amp; Yaseen, I. (2024). Elevating Survivability in Next-Gen IoT-Fog-Cloud Networks: Scheduling Optimization With the Metaheuristic Mountain Gazelle Algorithm. IEEE Transactions on Consumer Electronics, 70(1), 3802–3809. https://doi.org/10.1109/TCE.2024.3371774</text>
                    <pub-id pub-id-type="doi">10.1109/TCE.2024.3371774</pub-id>
                  </element-citation>
      </ref>
            <ref id="ref9">
        <element-citation publication-type="journal">
          <text>Mahapatra, A., Majhi, S. K., Mishra, K., Pradhan, R., Rao, D. C., &amp; Panda, S. K. (2024). An Energy-Aware Task Offloading and Load Balancing for Latency-Sensitive IoT Applications in the Fog-Cloud Continuum. IEEE Access, 12, 14334–14349. https://doi.org/10.1109/ACCESS.2024.3357122</text>
                    <pub-id pub-id-type="doi">10.1109/ACCESS.2024.3357122</pub-id>
                  </element-citation>
      </ref>
            <ref id="ref10">
        <element-citation publication-type="journal">
          <text>Mammadov, S., &amp; Kucukkulahli, E. (2025). A User-Centric Smart Library System: IoT-Driven Environmental Monitoring and ML-Based Optimization with Future Fog–Cloud Architecture. Applied Sciences 2025, Vol. 15, Page 3792, 15(7), 3792. https://doi.org/10.3390/APP15073792</text>
                    <pub-id pub-id-type="doi">10.3390/APP15073792</pub-id>
                  </element-citation>
      </ref>
            <ref id="ref11">
        <element-citation publication-type="journal">
          <text>Marzouk, M., &amp; Atef, M. (2022a). Assessment of Indoor Air Quality in Academic Buildings Using IoT and Deep Learning. Sustainability 2022, Vol. 14, Page 7015, 14(12), 7015. https://doi.org/10.3390/SU14127015</text>
                    <pub-id pub-id-type="doi">10.3390/SU14127015</pub-id>
                  </element-citation>
      </ref>
            <ref id="ref12">
        <element-citation publication-type="journal">
          <text>Marzouk, M., &amp; Atef, M. (2022b). Assessment of Indoor Air Quality in Academic Buildings Using IoT and Deep Learning. Sustainability 2022, Vol. 14, Page 7015, 14(12), 7015. https://doi.org/10.3390/SU14127015</text>
                    <pub-id pub-id-type="doi">10.3390/SU14127015</pub-id>
                  </element-citation>
      </ref>
            <ref id="ref13">
        <element-citation publication-type="journal">
          <text>Sadick, A. M., Kpamma, Z. E., &amp; Agyefi-Mensah, S. (2020). Impact of indoor environmental quality on job satisfaction and self-reported productivity of university employees in a tropical African climate. Building and Environment, 181, 107102. https://doi.org/10.1016/J.BUILDENV.2020.107102</text>
                    <pub-id pub-id-type="doi">10.1016/J.BUILDENV.2020.107102</pub-id>
                  </element-citation>
      </ref>
            <ref id="ref14">
        <element-citation publication-type="journal">
          <text>Salamone, F., Bellazzi, A., Belussi, L., Damato, G., Danza, L., Dell’aquila, F., Ghellere, M., Megale, V., Meroni, I., &amp; Vitaletti, W. (2020). Evaluation of the Visual Stimuli on Personal Thermal Comfort Perception in Real and Virtual Environments Using Machine Learning Approaches. Sensors 2020, Vol. 20, Page 1627, 20(6), 1627. https://doi.org/10.3390/S20061627</text>
                    <pub-id pub-id-type="doi">10.3390/S20061627</pub-id>
                  </element-citation>
      </ref>
            <ref id="ref15">
        <element-citation publication-type="journal">
          <text>Shah, S. K., Tariq, Z., Lee, J., &amp; Lee, Y. (2020). Real-Time Machine Learning for Air Quality and Environmental Noise Detection. Proceedings - 2020 IEEE International Conference on Big Data, Big Data 2020, 3506–3515. https://doi.org/10.1109/BIGDATA50022.2020.9377939</text>
                    <pub-id pub-id-type="doi">10.1109/BIGDATA50022.2020.9377939</pub-id>
                  </element-citation>
      </ref>
            <ref id="ref16">
        <element-citation publication-type="journal">
          <text>Tabuenca, B., Borner, D., &amp; Kalz, M. (2021). Effects of an Ambient Learning Display on Noise Levels and Perceived Learning in a Secondary School. IEEE Transactions on Learning Technologies, 14(1), 69–80. https://doi.org/10.1109/TLT.2021.3056038</text>
                    <pub-id pub-id-type="doi">10.1109/TLT.2021.3056038</pub-id>
                  </element-citation>
      </ref>
            <ref id="ref17">
        <element-citation publication-type="journal">
          <text>Twardella, D., Matzen, W., Lahrz, T., Burghardt, R., Spegel, H., Hendrowarsito, L., Frenzel, A. C., &amp; Fromme, H. (2012). Effect of classroom air quality on students’ concentration: Results of a cluster-randomized cross-over experimental study. Indoor Air, 22(5), 378–387. https://doi.org/10.1111/J.1600-0668.2012.00774.X;</text>
                    <pub-id pub-id-type="doi">10.1111/J.1600-0668.2012.00774.X;</pub-id>
                  </element-citation>
      </ref>
            <ref id="ref18">
        <element-citation publication-type="journal">
          <text>Umenga Weerasuriya, A., Wai Tsang, K., Li, Y., Lee, C.-C., Aflaki, A., Esfandiari, M., &amp; Jarrahi, A. (2023). Multi-Criteria Evaluation of a Library’s Indoor Environmental Quality in the Tropics. Buildings 2023, Vol. 13, Page 1233, 13(5), 1233. https://doi.org/10.3390/BUILDINGS13051233</text>
                    <pub-id pub-id-type="doi">10.3390/BUILDINGS13051233</pub-id>
                  </element-citation>
      </ref>
            <ref id="ref19">
        <element-citation publication-type="journal">
          <text>Viola, A. U., James, L. M., Schlangen, L. J. M., &amp; Dijk, D. J. (2008). Blue-enriched white light in the workplace improves self-reported alertness, performance and sleep quality. Scandinavian Journal of Work, Environment and Health, 34(4), 297–306. https://doi.org/10.5271/SJWEH.1268</text>
                    <pub-id pub-id-type="doi">10.5271/SJWEH.1268</pub-id>
                  </element-citation>
      </ref>
            <ref id="ref20">
        <element-citation publication-type="journal">
          <text>Zareb, M., Bakhti, B., Bouzid, Y., Batista, C. E., Ternifi, I., &amp; Abdenour, M. (2021). An intelligent IoT fuzzy based approach for automated indoor air quality monitoring. 2021 29th Mediterranean Conference on Control and Automation, MED 2021, 770–775. https://doi.org/10.1109/MED51440.2021.9480313</text>
                    <pub-id pub-id-type="doi">10.1109/MED51440.2021.9480313</pub-id>
                  </element-citation>
      </ref>
          </ref-list>
  </back>
</article>
