Student or educator? Sign up with your school email and get 6 months of Pro free.

The DOI matches its Crossref record; 40 of the 40 references it cites were found in a registry or an index; 3 of the 8 sources its claims cite were checked for support.

Sensor-Based Early Activity Recognition Inside Buildings to Support Energy and Comfort Management Systems

Open access

Abstract

Building Energy and Comfort Management (BECM) systems have the potential to considerably reduce costs related to energy consumption and improve the efficiency of resource exploitation, by implementing strategies for resource management and control and policies for Demand-Side Management (DSM). One of the main requirements for such systems is to be able to adapt their management decisions to the users’ specific habits and preferences, even when they change over time. This feature is fundamental to prevent users’ disaffection and the gradual abandonment of the system. In this paper, a sensor-based system for analysis of user habits and early detection and prediction of user activities is presented. To improve the resulting accuracy, the system incorporates statistics related to other relevant external conditions that have been observed to be correlated (e.g., time of the day). Performance evaluation on a real use case proves that the proposed system enables early recognition of activities after only 10 sensor events with an accuracy of 81 % . Furthermore, the correlation between activities can be used to predict the next activity with an accuracy of about 60 % .

Overview

CoChat Summary

TL;DR

The paper presents a sensor-based system for early recognition and prediction of user activities inside buildings to support Building Energy and Comfort Management. On the Aruba dataset, it reports early activity recognition after 10 sensor events with 81% accuracy and next-activity prediction with about 60% accuracy.

Key findings

  • Adding a statistics-based filter improved average activity recognition accuracy from 70.43% without the filter to 81.14% with an observation window of W = 10 sensor events.
  • Accuracy varied with observation window size: W = 15 gave 87.02%, W = 20 gave 84.09%, and W = 10 gave 81.14%, with 95% confidence intervals of ±1.56%, ±1.77%, and ±1.43%, respectively.
  • Prediction of subsequent activities reached about 60% accuracy, and transition/conditional probability tables were used to show correlations between activities.
  • Training from 2 weeks to 3 months increased overall accuracy, but the gain from 2 months to 3 months was less prominent; 95% confidence intervals ranged from ±1.95% for 2 weeks to ±1.68% for 3 months.

Methods

  • The system was tested on the Aruba real-world dataset from the CASAS smart environment project, using one smart apartment with motion, contact, and temperature sensors and one resident; 10 activities were considered, with a training time of 2 months, one week of test data, and W = 10 sensor events.
  • Training computes feature vectors from rates of event occurrences per sensor within activity time windows and model vectors as mean rates per activity; it also stores statistics G_j about relevant external conditions, such as average starting and ending times and their standard deviations.
  • The running phase uses sensor-based windowing, a Naïve Bayes Classifier, filtering based on G_j statistics, and cosine similarity between model and observed feature vectors to label the most likely activity; performance is evaluated by classification accuracy, confusion matrices, and activity transition/conditional probabilities.

Limitations

  • The paper states that the fixed observation window size creates a tradeoff: wider windows better recognize long activities but require more time before assigning an activity, while smaller windows can confuse similar activities.
  • The paper states that longer training periods improve accuracy but cause longer waiting times, and that retraining may be needed to detect changes in user habits.
  • The paper states that the approach was designed and tested only for consecutive activities performed by a single user; extension to multiple residents and concurrent or interleaved activities would require changes. Evident from the text: the dataset contains only one resident, and unlabeled “Other activity” events were ignored.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3390/en12132631

Cites

40 of 40 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

1 Valid · 6 Not valid · 3 Not checked · 57 Own results and interpretations

25 more claims

Create a free account to see all 67 claims and every source verdict.

See all claims — free account
  • Probably not a substantial claim. 26
  • It may state two separate claims as one. 15

Create a free account to see these claims.

Citations

Cites

Show all 40 references

Research works printed but not linked: 2

  1. Rohini, 2014, A New Approach to Behavioral Reasoning in Smart Homes using DVSM Algorithm, Int. J. Adv. Res. Comput. Sci. Softw. Eng., 4, 334
  2. Kaur, 2014, A review article on Naive Bayes classifier with various smoothing techniques, Int. J. Comput. Sci. Mob. Comput., 3, 864

Cited by

See all 13

Metadata

Authors
Marcello, F., Pilloni, V., & Giusto, D.
Affiliations
University of Cagliari (IT)
Corresponding author
Pilloni, Virginia
Venue
Energies
Publisher
MDPI AG
Published
2019-07-09
Volume
12
Issue
13
Pages
2631
ISSN
1996-1073 (ISSN-L)
OpenAlex
W2956411813
Cited by
13
References
40
Citations by year
2026: 1 · 2025: 1 · 2024: 3 · 2021: 4 · 2020: 2 · 2019: 2
FWCI
0.90
Type
journal article
Language
en
Open access
Locations
  • Energies (publishedVersion, cc-by)
  • RePEc: Research Papers in Economics (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • UNICA IRIS Institutional Research Information System (University of Cagliari) (submittedVersion, cc-by)
Funders
  • Ministero dell’Istruzione, dell’Università e della Ricerca (SCN 00281)
  • Ministero dello Sviluppo Economico (PON 2014/2020 POS. 395)
  • Regione Autonoma della Sardegna (L.R. 7/2007 annuity 2015, Grant Agreement no. 68)
  • Fondazione Banco di Sardegna (L.R. 7/2007 annuity 2017 - DGR 28/21 of 17/05/2015)
Topics
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Context-Aware Activity Recognition Systems
  • Physical Sciences › Engineering › Building and Construction › Building Energy and Comfort Optimization
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Smart Grid Energy Management
Keywords
activity recognition, building energy management, demand-side management, activity prediction, indoor occupancy detection

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-05, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-05 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Sensor-Based Early Activity Recognition Inside Buildings to Support Energy and Comfort Management Systems. https://app.cochat.ai/papers/10.3390/en12132631
The paper
Marcello, F., Pilloni, V., & Giusto, D. (2019). Sensor-Based Early Activity Recognition Inside Buildings to Support Energy and Comfort Management Systems. Energies. https://doi.org/10.3390/en12132631

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.