pith. sign in

arxiv: 1907.10818 · v1 · pith:NWVFD4VLnew · submitted 2019-06-20 · 📡 eess.SP · cs.LG

On Mining IoT Data for Evaluating the Operation of Public Educational Buildings

Na Zhu , Aris Anagnostopoulos , Ioannis Chatzigiannakis This is my paper

Pith reviewed 2026-05-25 19:07 UTC · model grok-4.3

classification 📡 eess.SP cs.LG
keywords IoTdata miningenergy efficiencyeducational buildingssensor databuilding operationssustainable management
0
0 comments X

The pith

Data mining on two years of IoT sensor readings from 18 school buildings yields operational changes that lower energy use.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper collects sensor data over two years from IoT devices installed across 18 public school buildings in Greece, Italy and Sweden. It then applies data mining techniques to examine interactions between indoor and outdoor conditions and overall building performance. The resulting analysis is intended to supply building managers and custodial staff with concrete guidance on adjusting daily operations. A sympathetic reader cares because public educational systems run thousands of buildings whose energy costs and environmental impact depend on practical management decisions. The evaluation concludes that the mined patterns can directly support measures that reduce a building's energy footprint.

Core claim

Applying data mining to the two-year IoT sensor dataset from 18 buildings across three countries produces actionable insights that let managers and staff adjust operations in ways that measurably lower energy consumption.

What carries the argument

Data mining applied to multi-year indoor-outdoor sensor streams collected from 18 school buildings.

If this is right

  • Building managers receive quantitative evidence for adjusting heating, ventilation and occupancy schedules.
  • Custodial staff can prioritize maintenance actions that directly affect energy use.
  • Public educational authorities obtain a repeatable method for evaluating performance across buildings of different ages and locations.
  • IoT infrastructure investments are justified by the ability to extract operationally useful patterns from the collected readings.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The same mining workflow could be tested on real-time streams to issue alerts when operations deviate from energy-efficient patterns.
  • Results from the three-country sample raise the question of whether similar patterns appear in non-educational public buildings.
  • If the identified changes are implemented, follow-up metering could quantify the actual energy reduction achieved.

Load-bearing premise

The two-year sensor dataset from the 18 buildings is representative, sufficiently complete, and free of major quality issues that would prevent reliable extraction of generalizable operational recommendations.

What would settle it

Re-running the same mining pipeline on the dataset produces no recurring, actionable operational patterns that correspond to measured energy reductions when the suggested changes are implemented in the buildings.

Figures

Figures reproduced from arXiv: 1907.10818 by Aris Anagnostopoulos, Ioannis Chatzigiannakis, Na Zhu.

Figure 1
Figure 1. Figure 1: Educational Building specific IoT architecture [PITH_FULL_IMAGE:figures/full_fig_p003_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Data availability per School Building 5 [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Data availability per Sensor Type 7 [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Temperature sensor time series processing [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Indoor temperature histogram for three classrooms during Sep/17 to Oct/17 [PITH_FULL_IMAGE:figures/full_fig_p009_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Site thermal comfort during Sep/17 to Oct/17 [PITH_FULL_IMAGE:figures/full_fig_p010_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Classroom temperature during 30/Sep (Saturday) [PITH_FULL_IMAGE:figures/full_fig_p011_7.png] view at source ↗
read the original abstract

Public educational systems operate thousands of buildings with vastly different characteristics in terms of size, age, location, construction, thermal behavior and user communities. Their strategic planning and sustainable operation is an extremely complex and requires quantitative evidence on the performance of buildings such as the interaction of indoor-outdoor environment. Internet of Things (IoT) deployments can provide the necessary data to evaluate, redesign and eventually improve the organizational and managerial measures. In this work a data mining approach is presented to analyze the sensor data collected over a period of 2 years from an IoT infrastructure deployed over 18 school buildings spread in Greece, Italy and Sweden. The real-world evaluation indicates that data mining on sensor data can provide critical insights to building managers and custodial staff about ways to lower a building's energy footprint through effectively managing building operations.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 0 minor

Summary. The paper presents a data mining approach to analyze two years of sensor data from an IoT infrastructure deployed across 18 school buildings in Greece, Italy, and Sweden. It claims that this analysis yields critical insights for building managers and custodial staff on operational adjustments that can lower a building's energy footprint, based on a real-world evaluation of the collected data.

Significance. If the central claim holds with supporting quantitative evidence, the work could demonstrate a practical application of IoT sensor data mining for sustainable management of public educational buildings with varying characteristics. It would add to the literature on data-driven building operations by providing case-study insights from a multi-country deployment. However, the current presentation supplies only the existence of the dataset and an assertion of insights, without demonstrated links to energy outcomes.

major comments (2)
  1. [Abstract] Abstract: The real-world evaluation asserts that data mining 'can provide critical insights ... about ways to lower a building's energy footprint through effectively managing building operations,' yet supplies no before/after consumption figures, no controlled comparison between buildings that acted on versus ignored the patterns, and no estimated savings attributable to the mined rules. This leaves the central claim correlational rather than evidentiary.
  2. [Evaluation (implied in abstract)] The manuscript does not address the weakest assumption that the two-year sensor dataset from the 18 buildings is representative, sufficiently complete, and free of major quality issues that would prevent reliable extraction of generalizable operational recommendations.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed review and constructive criticism. We address each major comment below and outline revisions to strengthen the manuscript.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The real-world evaluation asserts that data mining 'can provide critical insights ... about ways to lower a building's energy footprint through effectively managing building operations,' yet supplies no before/after consumption figures, no controlled comparison between buildings that acted on versus ignored the patterns, and no estimated savings attributable to the mined rules. This leaves the central claim correlational rather than evidentiary.

    Authors: The referee is correct that the evaluation identifies operational patterns from sensor data but does not include before/after energy consumption measurements, controlled comparisons, or quantified savings. The manuscript's primary contribution is the data mining methodology applied to a multi-country IoT deployment and the extraction of patterns that building managers could act upon. We will revise the abstract and evaluation sections to clarify that the insights indicate potential opportunities for energy reduction based on observed correlations, rather than claiming demonstrated savings. Direct attribution of savings would require a separate intervention study, which is outside the scope of this work. revision: yes

  2. Referee: [Evaluation (implied in abstract)] The manuscript does not address the weakest assumption that the two-year sensor dataset from the 18 buildings is representative, sufficiently complete, and free of major quality issues that would prevent reliable extraction of generalizable operational recommendations.

    Authors: We agree that explicit discussion of data quality, completeness, and representativeness is needed to support the reliability of the extracted recommendations. The deployment spans diverse buildings across three countries over two years, but the current text does not detail preprocessing steps or limitations. We will add a dedicated subsection on data quality assessment, missing value handling, and caveats regarding generalizability in the revised manuscript. revision: yes

Circularity Check

0 steps flagged

No significant circularity; empirical data-mining study without derivations or self-referential claims

full rationale

The paper applies standard data-mining techniques to a two-year IoT sensor dataset from 18 buildings and asserts that the resulting patterns supply operational insights for energy management. No equations, model fits, predictions, uniqueness theorems, or ansatzes appear. The central claim is an empirical observation about the utility of the mined patterns rather than a quantity derived from prior results by the same authors. No load-bearing self-citation chain or definitional loop is present; the work is self-contained as a descriptive case study.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

The work is an applied empirical study; the abstract introduces no mathematical derivations, free parameters, background axioms, or new postulated entities.

pith-pipeline@v0.9.0 · 5671 in / 927 out tokens · 22667 ms · 2026-05-25T19:07:44.157495+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

26 extracted references · 26 canonical work pages

  1. [1]

    Technical report, US Environmental Protection Agency (EPA), Washington DC, USA, 2008

    Energy star building upgrade manual. Technical report, US Environmental Protection Agency (EPA), Washington DC, USA, 2008

    work page 2008

  2. [2]

    Akribopoulos, I

    O. Akribopoulos, I. Chatzigiannakis, C. Koninis, and E. Theodoridis. A web services-oriented architecture for integrating small programmable objects in the web of things. In Proceedings of the 2010 Developments in E-systems Engineering , DESE ’10, pages 70–75, Washington, DC, USA,

    work page 2010

  3. [3]

    IEEE Computer Society

    work page

  4. [4]

    Akrivopoulos, N

    O. Akrivopoulos, N. Zhu, D. Amaxilatis, C. Tselios, A. Anagnostopoulos, and I. Chatzigiannakis. A fog computing-oriented, highly scalable iot framework for monitoring public educational buildings. In 2018 IEEE International Conference on Communications (ICC) , pages 1–6. IEEE, 2018. 11

    work page 2018

  5. [5]

    Allab, M

    Y. Allab, M. Pellegrino, X. Guo, E. Nefzaoui, and A. Kindinis. Energy and comfort assessment in educational building: Case study in a french university campus. Energy and Buildings , 143(Supple- ment C):202 – 219, 2017

    work page 2017

  6. [6]

    Amaxilatis, O

    D. Amaxilatis, O. Akrivopoulos, I. Chatzigiannakis, and C. Tselios. Enabling stream processing for people-centric iot based on the fog computing paradigm. In 2017 22nd IEEE International Conference on Emerging Technologies and Factory Automation (ETF A) , pages 1–8. IEEE, 2017

    work page 2017

  7. [7]

    Amaxilatis, O

    D. Amaxilatis, O. Akrivopoulos, G. Mylonas, and I. Chatzigiannakis. An iot-based solution for monitoring a fleet of educational buildings focusing on energy efficiency. Sensors, 17(10), 2017

    work page 2017

  8. [8]

    Standard 55 – Thermal Environmental Conditions for Human Occupancy

    ASHRAE. Standard 55 – Thermal Environmental Conditions for Human Occupancy. https://www. ashrae.org/resources--publications/bookstore/standard-55-and-user-s-manual . [On- line]

    work page

  9. [9]

    Bak-Bir, D

    Z. Bak-Bir, D. Clements-Croome, N. Kochhar, H. Awbi, and M. Williams. Ventilation rates in schools and pupils performance. Building and Environment , 48(Supplement C):215 – 223, 2012

    work page 2012

  10. [10]

    Baumgartner, I

    T. Baumgartner, I. Chatzigiannakis, M. Danckwardt, C. Koninis, A. Kr¨ oller, G. Mylonas, D. Pfis- terer, and B. Porter. Virtualising testbeds to support large-scale reconfigurable experimental facil- ities. In J. S. Silva, B. Krishnamachari, and F. Boavida, editors, Wireless Sensor Networks , pages 210–223, Berlin, Heidelberg, 2010. Springer Berlin Heidelberg

    work page 2010

  11. [11]

    Bottaccioli, A

    L. Bottaccioli, A. Aliberti, F. Ugliotti, E. Patti, A. Osello, E. Macii, and A. Acquaviva. Building energy modelling and monitoring by integration of iot devices and building information models. In 2017 IEEE 41st Annual Computer Software and Applications Conference (COMPSAC) , volume 1, pages 914–922, July 2017

    work page 2017

  12. [12]

    Chatzigiannakis, H

    I. Chatzigiannakis, H. Hasemann, M. Karnstedt, O. Kleine, A. Kroller, M. Leggieri, D. Pfisterer, K. Romer, and C. Truong. True self-configuration for the IoT. In 2012 3rd IEEE International Conference on the Internet of Things , pages 9–15. IEEE, Oct. 2012

    work page 2012

  13. [13]

    Chatzigiannakis, A

    I. Chatzigiannakis, A. Kinalis, and S. Nikoletseas. An adaptive power conservation scheme for heterogeneous wireless sensor networks with node redeployment. In Proceedings of the Seventeenth Annual ACM Symposium on Parallelism in Algorithms and Architectures , SPAA ’05, pages 96–105, New York, NY, USA, 2005. ACM

    work page 2005

  14. [14]

    Chatzigiannakis, G

    I. Chatzigiannakis, G. Mylonas, and S. Nikoletseas. Modeling and evaluation of the effect of obstacles on the performance of wireless sensor networks. In 39th Annual Simulation Symposium (ANSS’06) , pages 11 pp.–60, April 2006

    work page 2006

  15. [15]

    Chatzigiannakis, G

    I. Chatzigiannakis, G. Mylonas, and S. E. Nikoletseas. jwebdust : A java-based generic application environment for wireless sensor networks. In DCOSS, volume 3560 of Lecture Notes in Computer Science, pages 376–386. Springer, 2005

    work page 2005

  16. [16]

    Crosby and A

    K. Crosby and A. B. Metzger. Powering down: A toolkit for behavior-based energy conservation in k-12 schools. Technical report, U.S. Green Building Council (USGBC), Washington DC, USA, 2012

    work page 2012

  17. [17]

    J. E. Cross. Shifting organizational culture - innovation for energy conservation. Technical report, Garrison Institute, New York, USA, 2012

    work page 2012

  18. [18]

    T. Hoyt, S. Schiavon, A. Piccioli, D. Moon, and K. Steinfeld. CBE Thermal Comfort Tool. http:// cbe.berkeley.edu/comforttool/, 2013. Center for the Built Environment, University of California Berkeley

    work page 2013

  19. [19]

    Karthikeyan and P

    S. Karthikeyan and P. T. V. Bhuvaneswari. Iot based real-time residential energy meter monitoring system. In 2017 Trends in Industrial Measurement and Automation (TIMA) , pages 1–5, Jan 2017

    work page 2017

  20. [20]

    Miller and F

    C. Miller and F. Meggers. Mining electrical meter data to predict principal building use, performance class, and operations strategy for hundreds of non-residential buildings. Energy and Buildings , 156(Supplement C):360 – 373, 2017. 12

    work page 2017

  21. [21]

    Mylonas, D

    G. Mylonas, D. Amaxilatis, I. Chatzigiannakis, A. Anagnostopoulos, and F. Paganelli. Enabling sustainability and energy awareness in schools based on iot and real-world data. IEEE Pervasive Computing, 17(4):53–63, 2018

    work page 2018

  22. [22]

    Pocero, D

    L. Pocero, D. Amaxilatis, G. Mylonas, and I. Chatzigiannakis. Open source iot meter devices for smart and energy-efficient school buildings. HardwareX, 2017

    work page 2017

  23. [23]

    Schelly, J

    C. Schelly, J. E. Cross, W. S. Franzen, P. Hall, and S. Reeve. Reducing energy consumption and creating a conservation culture in organizations: A case study of one public school district. Environment & Behavior , 43(3):316–343, May 2011

    work page 2011

  24. [24]

    Schelly, J

    C. Schelly, J. E. Cross, W. S. Franzen, P. Hall, and S. Reeve. How to go green: Creating a conservation culture in a public high school through education, modeling, and communication. Journal of Environmental Education , 43(3):143–161, 2012

    work page 2012

  25. [25]

    National Best Practices Manual For Build- ing High Performance Schools

    US Department of Energy. National Best Practices Manual For Build- ing High Performance Schools. https://www.energy.gov/eere/downloads/ national-best-practices-manual-building-high-performance-schools . [Online]

    work page

  26. [26]

    N. Zhu, A. Anagnostopoulos, and I. Chatzigiannakis. On mining iot data for evaluating the operation of public educational buildings. In 2018 IEEE International Conference on Pervasive Computing and Communications Workshops (PerCom Workshops) , pages 278–283, March 2018. 13

    work page 2018