Application-independent feature construction based on almost-closedness properties

Selmaoui-Folcher, Nazha; Boulicaut, Jean-François

doi:10.1007/s10115-010-0369-x

Cited by 5 publications

(2 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this paper, we show the interest of data mining methods to classify satellite images database. The first important point is the processing of a large database of images with For that, we propose to investigate a supervised classification method based on the well-known noisy tolerant patterns [3]. A ground truth mission will have to be conducted in a few Tuamotu atolls in order to assess more precisely the performances of our method.…”

Section: Discussionmentioning

confidence: 99%

Coconut fields classification using data mining on a large database of high-resolution Ikonos images

Desmier

Flouvat

Stoll

et al. 2011

2011 Sixth International Conference on Digital Information Management

Self Cite

View full text Add to dashboard Cite

Supervised classification of satellite images is a commonly used technique in Remote Sensing. It allows the production of thematic maps based on a training set chosen by domain experts. These training sets, called ROI (Regions Of Interest), statistically characterize each class (e.g. coconut, sand) of the satellite image. Thus, a set of ROI is manually created by domain expert for each image. When a large number of images with high resolution occurs, manual creation of ROI for each image can be very time and money consuming. In this paper, we propose a semi-automatic approach based on clustering to limit the number of ROI done by experts. Then, we use decision trees on a binary decomposition of RGB components to improve the classification. Experiments have been done on 306 high resolution images of Tuamotu archipelago.

show abstract

Section: Discussionmentioning

confidence: 99%

Coconut fields classification using data mining on a large database of high-resolution Ikonos images

Desmier

Flouvat

Stoll

et al. 2011

2011 Sixth International Conference on Digital Information Management

Self Cite

View full text Add to dashboard Cite

show abstract

“…Other work has considered the creation of derived datasets based on the presence (or absence) of particular patterns within the data. For example, Gay et al (2012) propose a method suitable for classification of data with binary valued attributes. Their work differs from the method proposed in this paper which deals with real valued meter readings and includes a symbolisation step to handle approximate matching of patterns.…”

Section: Motifsmentioning

confidence: 99%

A Method for Evaluating Options for Motif Detection in Electricity Meter Data

Dent¹,

Craig²,

Aickelin³

et al. 2021

Journal of Data Science

View full text Add to dashboard Cite

Investigation of household electricity usage patterns, and matching the patterns to behaviours, is an important area of research given the centrality of such patterns in addressing the needs of the electricity industry. Additional knowledge of household behaviours will allow more effective targeting of demand side management (DSM) techniques. This paper addresses the question as to whether a reasonable number of meaningful motifs, that each represent a regular activity within a domestic household, can be identified solely using the household level electricity meter data. Using UK data collected from several hundred households in Spring 2011 monitored at a frequency of five minutes, a process for finding repeating short patterns (motifs) is defined. Different ways of representing the motifs exist and a qualitative approach is presented that allows for choosing between the options based on the number of regular behaviours detected (neither too few nor too many).

show abstract