Using Closed n-set Patterns for Spatio-Temporal Classification - Archive ouverte HAL Access content directly
Conference Papers Year : 2014

Using Closed n-set Patterns for Spatio-Temporal Classification

Abstract

Today,huge volumes of sensor data are collected from many different sources. One of the most crucial data mining task considering this data is the ability to predict and classify data to anticipate trends or failures and take adequate steps. While the initial data might be of limited interest itself, the use of additional information, e.g., latent attributes, spatio-temporal details, etc., can add significant values and interestingness. In this paper we present a classification approach, called Closed n-set Spatio-Temporal Classification (CnSC), which is based on the use of latent attributes, pattern mining, and classification model construction. As the amount of patterns generated is huge, we employ a scalable NoSQL-based graph database for efficient storage and retrieval. The classification model for a specific context is constructed by aggre- gating the most similar patterns. By considering hierarchies in the latent attributes, we define pattern and context similarity scores, and use these for ranking and aggregation. Pattern-based classification shows compet- itive results compared with other prediction strategies
Fichier principal
Vignette du fichier
samelecivius_15178.pdf (983.79 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-01399862 , version 1 (21-11-2016)

Identifiers

  • HAL Id : hal-01399862 , version 1
  • OATAO : 15178

Cite

Saulius Samulevicius, Yoann Pitarch, Torben Bach Pedersen. Using Closed n-set Patterns for Spatio-Temporal Classification. 16th International Conference on Data Warehousing and Knowledge Discovery (DaWaK 2014), Sep 2014, Munich, Germany. pp. 274-287. ⟨hal-01399862⟩
68 View
99 Download

Share

Gmail Facebook X LinkedIn More