Development of Traffic Geo-Application Based on Big Data Processing

Keywords:
Hadoop, HDFS, HBase, HiveQL, Shema on Read.

Abstract:
Past is a prelude. Historical data provides signals that indicate what may happen in the future. Analytic tools can be especially useful in helping travel and transportation companies mine and refine data to determine which information is valuable for optimizing business outcomes. In this research we define a methodology for developing traffic geo-applications, based on: 1) relational data model for calculated predefined attributes of the maps, 2) wide column data model for source traffic data store, and 3) schema on read modeling approach for traffic data analysis, ie. calculating of the attributes of the maps. According to the proposed methodology we have developed Traffic Counting geo-application, which enables on-line access, displaying and geo-location of traffic indicators, calculated and stored on Apache Hadoop database. We concluded that our approach is appropriate in scenarios that involve batch Big Data processing and predefined data analysis, but not appropriate for the on-line ad-hoc definition of arbitrary queries.

CITATION:

IEEE format

Copy to clipboard

S. Janković, S. Mladenović, A. Uzelac, S. Zdravković, S. Aćimović, “Development of Traffic Geo-Application Based on Big Data Processing,” in Sinteza 2017 - International Scientific Conference on Information Technology and Data Related Research, Belgrade, Singidunum University, Serbia, 2017, pp. 458-463. doi:10.15308/Sinteza-2017-458-463

APA format

Copy to clipboard

Janković, S., Mladenović, S., Uzelac, A., Zdravković, S., Aćimović, S. (2017). Development of Traffic Geo-Application Based on Big Data Processing. Paper presented at Sinteza 2017 - International Scientific Conference on Information Technology and Data Related Research. doi:10.15308/Sinteza-2017-458-463

BibTeX format
Download

RefWorks Tagged format
Download