Sei sulla pagina 1di 6

Intelligent Urban Transport Decision Analysis

System Based on Mining in Big Data Analytics


and Data Warehouse

Khaoula Addakiri1, Hajar Khallouki2(&), and Mohamed Bahaj2


1
Computer Science Department, Ibn Zohr University, Ouarzazate, Morocco
2
Mathematics and Computer Science Department, Faculty of Sciences
and Technologies, Hassan I University, Settat, Morocco
Hajar.khallouki@gmail.com

Abstract. This paper conduct a study on the augmentation of the current


capabilities of the intelligent urban mobility and road transport in terms of the
analytics dimension focusing on the data mining and big data analytics
methodologies. A federated or a hybrid approach leverages the strengths and
mitigates the weaknesses of both data warehouse and big data analytics. We
discuss the challenges, requirements, integrated models, components, scenarios
and proposed solutions to the performance, efficiency, availability, security and
privacy concerns in the context of smart cities. Our approach relies on several
layers that run in parallel to collect and manage all collected data and create
several scenarios that will be used to assist urban mobility. The data warehouse
and big data analytics can serve as means to support clustering, classification,
recommending systems, frequent item set mining. The challenge here is to
populate the repository architecture with the schema, view definitions, metadata
and specify/integrate the types of this architecture (Centralized Metadata
repository, Distributed Metadata repository, Federated or Hybrid Metadata
repository).

Keywords: ITS  Urban mobility  Big data analytics  Data mining  Data
warehouse

1 Scope and Outline of the Paper

Plenty of solutions have been applied to mitigate and predict the number of road
accidents. Like establishing stringent rules and regulations. But most of them failed to
decrease road accidents. Many researchers work in the area of pervasive and context-
aware computing has developed different kinds of context models. [4] approach con-
tains different levels that represent the process of multimedia documents adaptation. [5]
allows the production of medical, genetic and scientific data between health profes-
sionals, scientists and patients.
In [6] the author designs intelligent transport organization optimization schemes
and analyzes the types of data processed on big data platform.
This paper attempts to integrate the servers, storage handling, knowledge man-
agement data and client tools for making decision. We provide an intelligent system for

© Springer Nature Switzerland AG 2020


M. Ezziyyani (Ed.): AI2SD 2019, LNNS 92, pp. 179–184, 2020.
https://doi.org/10.1007/978-3-030-33103-0_18
180 K. Addakiri et al.

accident prevention and detection for human life safety. The prevention part involves
the following aspects:
– Conduct surveys to analyze the driver behavior;
– Establish a decision support model for the detection and identification of accident
factors;
– Analyze data and determine trends in road accidents and identify potentially dan-
gerous accidents areas;
– Traffic flow forecasting;
– Emergency management;
– Urban mobility;
– Mitigating urban traffic congestion.

1.1 Big Data Analytics


The approach of Big data can support real-time access of metadata from source sys-
tems. It can also centrally and reliably maintain metadata definitions to the proper
locations of the accurate definitions in order to improve performance and availability.
This architecture proposed in [3] present a brief summary of the various modules in
Big data (Fig. 1).

Fig. 1. Big data modules

Large volumes of data sets derived from sophisticated sensors and social media
feeds are increasingly being used by the researchers. Processing a large amount of data
is not easy with conventional parallel computing, due to the failure of the compute
nodes and the scalability of the system. big data is a useful tool for improving the
performance and availability.
The 4V’s of big data – volume, velocity, variety and veracity makes the data
management and analytics challenging for the traditional data warehouses.
Intelligent Urban Transport Decision Analysis System 181

Large data analysis - the process of analyzing and exploring large data - can
generate operational and decisional information of a scale and specificity important for
the control of urban mobility and energy efficiency. The need to analyze and exploit
trend data collected by different sources is one of the main drivers of Big Data analytics
tools.
Big Data Analytics is rapidly evolving both in terms of functionality and the
underlying programming model. Such analytical functions support the integration of
results derived in parallel across distributed pieces of one or more data sources [2].

1.2 Data Warehouse


This paper focuses on the use of data warehouse as a supporting tool in decision
making in the context of smart cities. A data warehouse is a subject-oriented, inte-
grated, time-variant and non-volatile collection of data in support of management’s
decision making process Inmon [1].
The data warehouse gathers database scattered from different sources. We focus on
the benefits gained from using data warehouse.

2 Design and Architecture

The intelligent solution as it will be designed, will manage, analyze, alert, secure, and
improve the quality of services offered by data management systems in data Ware-
house, machine learning and Big Data analytics, these aspects allow the modeling of
urban mobility scenarios, Traffic flow forecasting and auditing of computations and
data.

2.1 Layer Acquisition, Cleaning, Loading Modeling Data


This view tries to provide a wide basis of integrated data and data modeling. These data
come from several heterogeneous sources (Raw data, CSV, XML files, JSON,
Sensors/embedded sensors, after the extracting phase, cleaning, filtering and the
determination the schema, view definitions, metadata).
In this layer, we present a brief summary of the security and privacy aspects of our
approach, the privacy requirements in urban mobility, and some of the existing privacy
solutions in urban mobility.
The preservation of privacy largely relies on technological limitations on the ability
to extract, analyze, and correlate potentially sensitive data sets. However, advances in
Big Data analytics provide tools to extract and utilize this data, making violations of
privacy easier. As a result, along with developing Big Data tools, it is necessary to
create safeguards to prevent abuse (Bryant, Katz, & Lazowska, 2008).
182 K. Addakiri et al.

Information Raw data


Videosurveillance
geographic Big Data
BigTable, Hbase,
Cassandra,
ZooKeeper
OLTP
Sensors/
embedded
sensors

CSV,XML,
JSON

Data
Data
Security
acquisition
and Privacy

2.2 Layer Processing, Querying and Data Analysis


The major challenge of the road safety of this paper comes from the innovative
character of the exploitation and the integration of some data associated to the field of
road traffic management, the optimization of its infrastructure, as well as the trends,
assets and constraints related to vehicles and pedestrians. Therefore, it is essential to
exploit, in real time, these data and all the infrastructure that manages them.
We aim to design an integrated and scalable architecture to access a shared pool of
configurable resources. This layer deals with the analysis, modeling and design of a
system for access management and real-time detection of anomalies.
The major requirements is to preserve the integrity, confidentiality or availability of
data in order to deduce recommendations on integrity of data-computations, and
correctness-freshness and support for decision-making.
The need to handle querying and data analysis from various applications and data
stores into the central repository may compromise data quality. We discuss how new
technologies can improve urban mobility and contribute to road safety and congestion
reduction in a smart city. Our approach is to gather, federate and synthesize data and
support a decision support system consisting of a set of recommendations.
Data Mining Algorithm Applied to Intelligent Transport System: Development of
data mining tools, and its integration into the part of the parallel processing of Big Data
within the Big Data analytics management system and its coexistence with data
warehouse.

2.3 Layer the Decision Support Technology


We discuss here how new technologies can improve urban mobility and contribute to
road safety and congestion reduction in a smart city. Our approach is to gather, federate
and synthesize data and support a decision support system consisting of a set of
recommendations.
Intelligent Urban Transport Decision Analysis System 183

Modeling the ITS organization as a complex system: The ITS organization


framework.

Decision Reports
making Analysis

Traffic flow forecasting/


congestion/accidents road/
emergency/
implementation of
visualization system for
vehicles and pedestrians

Machine
learning
Big data analytics

Pig, Hive,
Mahout and
RHadoop

OLAP

Datamining
Clustering/ NoSQL
Classification/ Databases/
Outlier Ware house ecosystem
Decision/ Hadoop
Visualization

Integration of
cloud computing ETL
and Internet of Ware house
Things:

Data
Processing

Big Data
BigTable, Hbase,
OLTP Cassandra,
ZooKeeper

3 Conclusion

We have established how data analytics could benefit ITS using scenarios including
efficient route guidance.
Regulations and policies of component organization in this architecture, and the
general availability of the components are not discussed in detail. Our first challenge is
the integration and conformity study of the different components of the proposed
architecture.
184 K. Addakiri et al.

References
1. Inmon, W.H.: Building the Data Warehouse, 2nd edn. Wiley, New York (1996)
2. El-Seoud, S.A., El-Sofany, H.F., Abdelfattah, M., Mohamed, R.: Big data and cloud
computing: trends and challenges. IJIM 11(2) (2017)
3. Ferandez, A., del Sara, R., López, V., Bawakid, A., del Jesus, M.J., Benitez, J.M., Herrera, F.:
Big data with cloud computing: an insight on the computing environment, mapreduce, and
programming frameworks. WIREs Data Min. Knowl. Discov. 4, 380–409 (2014). https://doi.
org/10.1002/widm.1134
4. Khallouki, H., Bahaj, M.: Context modeling architecture in pervasive computing environ-
ments for multimedia documents adaptation. In: 2016 5th International Conference on
Multimedia Computing and Systems (ICMCS), pp. 611–615. IEEE (2016)
5. Abatal, A., Khallouki, H., Bahaj, M.: A semantic smart interconnected healthcare system
using ontology and cloud computing. In: 2018 4th International Conference on Optimization
and Applications (ICOA), pp. 1–5. IEEE, April 2018
6. Ying, C.: Intelligent transport decision analysis system based on big data mining. In:
Advances in Computer Science Research (ACSR), vol. 73

Potrebbero piacerti anche