TU Berlin

Service-centric NetworkingJenke, M. (2016). Data Aggregation for Location Data Analytics. Bachelor Thesis. Technische Universität Berlin

Page Content

to Navigation

Bachelor Thesis: Data Aggregation for Location Data Analytics

Title: Data Aggregation for Location Data Analytics

Description: Over the last decade the emerging of a nearly omnipresent, accurate and robust network for mining spatiotemporal data was the fundamental key to the arising of several new services and applications in a wide range of domains. Spatiotemporal data contains the essential information that all these services are build upon. At the same time location data analytics basics are required for the usage, processing and visualization of the spatiotemporal data. Especially because in the very most cases these location data analytics will be set in a big data surrounding. Therefore one of the primary challenges is to verify a suitable data aggregation. Considering the high amount and frequency of incoming spatiotemporal data streams, an all-embracing storing and processing of all the received data is not suitable. Not only because of technical challenges but also due to numerous economical aspects and constraints. Aggregation and pre-aggregation are two tools that propose one possibility to dramatically decrease the amount and size of data streams while at the same time maintain the information in the best possible manner or to reduce the data to the core of the highest interest. Simultaneously a sophisticated data aggregation enables to not only adjust the size of a data set but furthermore to optimise the data structure for a later analysing, processing and query performance in regards of responsiveness and handling. It comes at a great challenge to keep the balance between aggregation on side and the loss of valuable information on the other side.

This thesis will introduce several approaches and aggregation strategies for spationtemporal data sets. The choice for a suitable aggregation strategy is based fundamentally on a successful requirement analysis, which considers both technical and business aspects. The later to be developed guidance is intended to help a potential user to succesfully select an aggregation strategy with regards to the user specific requirements. Afterwards the prototypical implementation of a developed framework is used to perform the selected aggregation strategies on the basis of a given real life data set and to evaluate the results in respect to quality and quantity.

Supervisor: Dr. Peter Ruppel

Type: Bachelor Thesis

Duration: 4 months

Navigation

Quick Access

Schnellnavigation zur Seite über Nummerneingabe