Selective Sampling for Sensor Type Classification in Buildings
A key barrier to applying any smart technology to a building is the requirement of locating and connecting to the necessary resources among the thousands of sensing and control points, i.e., the metadata mapping problem. Existing solutions depend on exhaustive manual annotation of sensor metadata --- a laborious, costly, and hardly scalable process. To reduce the amount of manual effort required, this paper presents a multi-oracle selective sampling framework to leverage noisy labels from information sources with unknown reliability such as existing buildings, which we refer to as weak oracles, for metadata mapping. This framework involves an interactive process, where a small set of sensor instances are progressively selected and labeled for it to learn how to aggregate the noisy labels as well as to predict sensor types. Two key challenges arise in designing the framework, namely, weak oracle reliability estimation and instance selection for querying. To address the first challenge, we develop a clustering-based approach for weak oracle reliability estimation to capitalize on the observation that weak oracles perform differently in different groups of instances. For the second challenge, we propose a disagreement-based query selection strategy to combine the potential effect of a labeled instance on both reducing classifier uncertainty and improving the quality of label aggregation. We evaluate our solution on a large collection of real-world building sensor data from 5 buildings with more than 11,000 sensors of 18 different types. The experiment results validate the effectiveness of our solution, which outperforms a set of state-of-the-art baselines.
- Research Organization:
- University of Virginia
- Sponsoring Organization:
- USDOE Office of Energy Efficiency and Renewable Energy (EERE), Energy Efficiency Office. Building Technologies Office
- DOE Contract Number:
- EE0008227
- OSTI ID:
- 1822657
- Report Number(s):
- DOE-UVA-0008227-7
- Journal Information:
- 2020 19th ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN), Conference: The 19th ACM/IEEE International Conference on Information Processing in Sensor Networks (IPSN), Sydney, NSW, Australia, 21-24 April 2020
- Country of Publication:
- United States
- Language:
- English
Active Learning with Statistical Models
|
journal | January 1996 |
Proactive learning
|
conference | January 2008 |
Plaster: an integration, benchmark, and development framework for metadata normalization methods
|
conference | November 2018 |
Automated point mapping for building control systems: Recent advances and future research needs
|
journal | January 2018 |
Efficient crowdsourcing of unknown experts using bounded multi-armed bandits
|
journal | September 2014 |
Scrabble: transferrable semi-automated semantic metadata normalization using intermediate representation
|
conference | November 2018 |
POEM: power-efficient occupancy-based energy management system
|
conference | January 2013 |
DeviceMien
|
conference | April 2019 |
Automated Metadata Construction to Support Portable Building Applications
|
conference | January 2015 |
Sentinel: occupancy based HVAC actuation using existing WiFi infrastructure within commercial buildings
|
conference | January 2013 |
A Bayesian Framework for Modeling Human Evaluations
|
conference | June 2015 |
Active learning with confidence-based answers for crowdsourcing labeling tasks
|
journal | November 2018 |
Maximum Likelihood Estimation of Observer Error-Rates Using the EM Algorithm
|
journal | January 1979 |
A Data-driven Meta-data Inference Framework for Building Automation Systems
|
conference | January 2015 |
Clustering-based Active Learning on Sensor Type Classification in Buildings
|
conference | October 2015 |
Brick: Towards a Unified Metadata Schema For Buildings
|
conference | January 2016 |
Towards Automatic Spatial Verification of Sensor Placement in Buildings
|
conference | November 2013 |
Towards automating the deployment of energy saving approaches in buildings
|
conference | November 2014 |
On Information and Sufficiency
|
journal | March 1951 |
Online crowdsourcing: Rating annotators and obtaining cost-effective labels
|
conference | June 2010 |
Truth inference in crowdsourcing
|
journal | January 2017 |
Zodiac
|
conference | November 2015 |
Mismatch string kernels for discriminative protein classification
|
journal | January 2004 |
A study of the effect of different types of noise on the precision of supervised learning techniques
|
journal | January 2010 |
Short Paper
|
conference | November 2015 |
A Bayesian Analysis of Some Nonparametric Problems
|
journal | March 1973 |
The Building Adapter: Towards Quickly Applying Building Analytics at Scale
|
conference | January 2015 |
Efficiently learning the accuracy of labeling sources for selective sampling
|
conference | January 2009 |
Comparison of linear correlation and a statistical dependency measure for inferring spatial relation of temperature sensors in buildings
|
conference | November 2014 |
The wisdom of minority
|
conference | January 2014 |
Strip, bind, and search: a method for identifying abnormal energy consumption in buildings
|
conference | January 2013 |
Combining Crowd and Expert Labels Using Decision Theoretic Active Learning
|
journal | September 2015 |
Similar Records
Scalable Pattern Matching in Metadata Graphs via Constraint Checking
The Building Adapter: Automatic Mapping of Commercial Buildings for Scalable Building Analytics