The ever-increasing road traffic collisions, improved gathering, analyzing and tackling constructive information in view of traffic safety has become one of most vital concerns for local traffic authorities. The ultimate goal is to guarantee sound traffic management and rapid access to updated road traffic information for city travelers. Therefore, an effort to devise a traffic safety information platform based on cloud computing technology, which serves as an outstanding channel for message between traffic authority, staff and the public. This platform relies on data gathering & storing using Big data methodologies, analyzing the data using Data warehousing, and mining model, eagerly personalized to present traffic environment and trustworthy data from traffic accidents. The platform is completely integrated with matching variables such as people, vehicles, and environment, weather and road conditions.
Keywords |
| data mining, cloud computing, data warehousing, clustering |
INTRODUCTION |
Overview of traffic safety |
| According to the statistics published by World Health Organization (WHO), annual road traffic deaths and injuries
around the world has reached 1.2 million and 50 million respectively, which regularly resulted in everlasting damages.
With the modern development of information technology and the eternally growing threat to life from road traffic
collisions, it has become one of most vital concerns for local traffic authorities to better gather, analyze and tackle
helpful information in view of traffic safety with a vision to ensure sound traffic management and for city travelers to
get speedy access to updated road traffic information in the expect of guiding their travel behaviors as well [2].
Therefore, an endeavor to design a traffic safety information management platform based on cloud computing, data
warehousing and data mining, which serves as an outstanding channel for communication between traffic authority,
staff and the public. It aims to obtain full benefit of road safety information database in effort to simplify, judge and
predict different safety information such as traffic accident black spots, vehicle types legally responsible to serious
accidents and weather based on already occurred accidents on one hand and share detailed traffic conditions without
hindrance such as road condition, potential risks, disasters and weather [21]. |
Data Warehouse |
| Data warehouse is a database used for reporting and data analysis. Integrating data from one or more contrasting
sources creates a middle storehouse of data. It store existing and historical data and are used for creating, trending
reports for senior management reporting such as yearly and quarterly comparisons. The data stored in the warehouse
is uploaded from the operational systems. The typical extract transform load (ETL) based data warehouse uses staging,
data integration, and access layers to house its key functions. The staging layer or staging database stores raw data
extracted from each of the dissimilar source data systems. The integration layer integrates the dissimilar data sets by
transforming the data from the staging layer often storing this transformed data in an operational data store (ODS)
database. The integrated data are then moved to yet another database, often called the data warehouse database, where
the data is arranged into hierarchical groups often called dimensions and into facts and aggregate facts. The access layer
helps users retrieve data. A data warehouse constructed from an integrated data source system does not require ETL,
staging databases, or operational data store databases. This integrated data warehouse architecture supports the drill
down from the aggregate data of the data warehouse to the transactional data of the integrated source data systems. The
data warehouse focuses on data storage. The main source of the data is cleaned, transformed, catalogued and made
available for use by managers and other business professionals for data mining, online analytical processing, market
research and decision support. However, the means to retrieve and analyze data, to extract, transform and load data, and to manage the data dictionary, business intelligence tools, tools to extract data into the repository, and tools to manage
and retrieve metadata are also considered important components of a data warehousing system. |
Data Mining |
| Data mining is the computational procedure of discovering patterns in huge data sets involving methods at the
intersection of artificial, machine learning, statistics, and database systems. The general objective of the data mining
process is to extract information from a data set and transform it into an understandable structure for further use. apart
from the raw analysis stage, it involves database & data management aspects, data pre-processing,
model and inference considerations, interestingness metrics, complexity considerations, post-processing of discovered
structures, visualization, and online updating. The real data mining task is the automatic or semi-automatic analysis of
large quantities of data to extract previously unfamiliar remarkable patterns such as groups of data records (cluster
analysis), unusual records (anomaly detection) and dependencies (association rule mining). These patterns can then be
seen as a kind of summary of the input data and may be used in further analysis. |
| Data mining involves six common classes of tasks |
| • Anomaly detection (Outlier/change/deviation detection) – The detection of unusual data records, that might be
interesting or data errors that require additional investigation. |
| • Association rule learning (Dependency modeling) – Searches for relationships between variables. |
| • Clustering – is the job of discovering groups and structures in the data that are in some way or another
"similar", without using recognized structures in the data. |
| • Classification – is the task of generalizing known structure to apply to new data. |
| • Regression – attempts to find a function which models the data with the least error. |
| • Summarization – providing a more dense representation of the data set, including visualization and report
generation. |
LITERATURE SURVEY |
| Hadoop [22] was created by Doug Cutting, the creator of Apache Lucene, the widely used text search library. Hadoop
has its origins in Apache Nutch, an open source web search engine, itself a part of the Lucene project. Mike Cafarella
and Doug Cutting estimated a system supporting a billion page index would cost around half a million dollars in
hardware, with a monthly running cost of $30,000. |
| Jana and Bandyopadhyay [3] analyzed several security threats and its measures, and suggested additional
recommendations on top of best practices adopted for Identity and Access Management(IAM) for mobile cloud users.
Pirker et al [4] presented privacy preserving cloud resource payment to enable mobile clients to anonymously consume
resources of a cloud service provider such that the provider is not able to track users' activity patterns. |
| Satyanarayanan [5] discussed a wide range of issues in areas such as privacy, software licensing, and business models
with the emergence of cloudlet based hardware/software ecosystem to support for example, cognitive assistance for
attention challenged mobile users, scalable crowd sourcing of first person video, and ubiquitous mobile access to one's
legacy world. |
| Abuelela and Olariu[6] envision VANETs as cloud computing networks as well. Vehicles will share the capability of
computing power, Internet access and storage to form conventional clouds. |
| Santos et al. [7] proposed a new platform to achieve trust in conventional clouds. Krautheim [8] also proposed a third
party to share the responsibility of security in cloud computing between the service provider and client, decreasing the
risk exposure to both. The bilinear aggregate signature has been extended to simultaneously audit multiple users. |
| Observing the fact that many devices (computing, sensing and storing devices) on vehicles are idle for a long while,
Olariu and his colleagues [9], [10], [11] proposed to share these devices as the computational engine of the cloud.
Ristenpart et al. [12] presented experiments of locating co-residence of other users in cloud virtual machines. |
| The mobile cloud application models are based on the standard cloud service model that includes Infrastructure as a
Service (IaaS) [14], Platform as a Service (PaaS) [15], and Software as a Service (SaaS) [16], [17]. Therefore, based on
the working of the application models, any of these service layers can be utilized. Some of the well known services for
mobile cloud computing include Amazon Elastic Compute Cloud (EC2) [18], GoogleApp Engine [15], and Microsoft
Azure [19]. |
THE TECHNOLOGIES AND CONCEPTS |
Cloud Computing |
| Cloud Computing can be used as a platform and one type of application program abreast of the latest computing
development such as distributed computing, parallel computing and grid computing. It uses a computer network to
offer computing resources such as data or software as a service to users who pay for it on demand. Thus, it breaks free
of the computing power and storage space margins of the traditional local computing model. User’s computers, mobile
phones, and other private devices might only contain an operation system and an Internet explorer, and they do not
have to to identify where the data is stored or who offer the software. Users present their computing tasks that cannot
be skilled by their local devices to the clouds. The clouds offer computing services and return the computing results to
them. |
Cloud computing benefits from several key characteristics |
| • Reliability The data is stored and the applications are running on the servers in the clouds. Users do not have
to worry about lost or corrupt data. |
| • Agility The clouds can distribute computing resources according to the user’s needs or preferences to provide
flexible management. |
| • Utility. Users do not have to buy expensive computing devices. They only need to pay for the computing
services provided by the clouds. |
| • Application programming interface (API) accessibility to software that enables machines to interact with cloud
software in the same way that a traditional user interface facilitates interaction between humans and
computers. Cloud computing systems typically use Representational State Transfer (REST) based APIs. |
| • Device and location independence facilitate users to access systems using a web browser despite of their
location or what device they use. As infrastructure is off site and accessed via the Internet, users can connect
from anywhere. |
| • Multitenancy enables sharing of resources and costs across a large pool of users thus allowing for
Centralization of infrastructure in locations with lower costs, Peak load capacity increases & Utilisation and
efficiency improvements for systems that are often only 10–20% utilised. |
| • Performance is monitored and consistent and loosely coupled architectures are constructed using web
services as the system interface. |
| • Security can get better due to centralization of data, improved security focused resources, etc., but concerns
can stick with about loss of control over definite perceptive data, and the lack of security for stored kernels. |
Analyzing the Data with Hadoop |
Hadoop |
| As infrastructure of distributed system, Users can explore the distributed programs and make full use of cluster of highspeed
process and storage without shrewd details about the distributed architecture substrates [22]. Hadoop is a group
of associated subprojects that descend beneath the umbrella of infrastructure for distributed computing. These projects
are hosted by the Apache Software Foundation, which provides support for a community of open source software
projects. To take advantage of the parallel processing that Hadoop provides, it is essential to express query as a Map
Reduce job. After some local, small-scale testing will be able to run it on a cluster of machines. |
Map and Reduce |
| Map Reduce is a programming model for data processing. The model is simple, yet not too simple to express useful
programs in. Map Reduce works by breaking the processing into two phases: the map phase and the reduce phase. Each
phase has key-value pairs as input and output, the types of which may be chosen by the programmer. The programmer
also specifies two functions: the map function and the reduce function. The input to our map phase is the raw data,
which choose a text input format that gives each line in the dataset as a text value. The key is the offset of the start of
the line from the beginning of the file. The map function is just a data preparation phase, setting up the data in such a
mode that the reducer function can do its work on it. The map function is also a good place to drop bad records. As can
be seen from the Fig. 1., it contains a lot of components. |
Avro |
| A data serialization system for efficient, cross language RPC, and relentless data storage. (At the time of this writing,
Avro had been created only as a new subproject, and no other Hadoop subprojects were using it yet.) |
HDFS |
| Hadoop consists of distributed filesystem called HDFS, which stands for Hadoop Distributed Filesystem. A distributed
filesystem that runs on huge clusters of product machines. HDFS is a filesystem intended for storing very large files
with streaming data access patterns, running on clusters on commodity hardware. |
Pig |
| A data flow language and execution environment for exploring very big datasets. Pig runs on HDFS and Map/Reduce
clusters. |
HBase |
| HBase is an open source, non-relational, distributed data base modeled after Google's BigTable and is written in Java.
HBase uses HDFS for its fundamental storage, and supports both batch style computations using Map/Reduce and
point queries (random reads).It is developed as part of Apache Software Foundation's Apache Hadoop project and runs
on apex of HDFS, providing BigTable like capabilities for Hadoop. |
ZooKeeper |
| A distributed, highly available coordination service. ZooKeeper provides primitive such as distributed locks that can be
used for building distributed applications. |
Hive and Mahout |
| Hive is a data warehouse platform based on Hadoop. By utilizing Hive, data extraction, transformation and loading can
easily realize. Furthermore, it’s potential to convert the program from QL to corresponding Map/reduce and base them
on Hadoop by HQL (Hive Query Language). Mahout is a distributed framework of machine learning and data mining
provide some traditional algorithm in the extensible ground of machine learning. |
PROBLEMS AND CONSTRUCTION MODE |
Problems |
| Public traffic information service has been in the interest worldwide. But, there is still a batch of weaknesses in this
area: |
| • Limited information service ability. |
| At present the majority information services rely on government and information gather, tackling &
releasing are self-governing to several extent. What’s added information content seems to be unsophisticated
and unusable. |
| • Drawback of service content, approach, eminence and range. |
| The present service content chiefly touches upon travel information and service approach is
considered to be obsolete and thus put down much room for development. |
| • Lack of frequently updated traffic safety information. |
| At present time, the significant traffic safety information is partial for information service provider &
it’s really sturdy congregation, storing and processing group information. |
| For the given problems above, an attempt to generate a traffic safety information platform based on cloud computing
for local traffic authorities and the public through establishing traffic safety information system. With the platform any
users can attach cloud service platform with their own terminal apparatus and keep them well educated of any traffic
safety information they are concerned about. |
The main research content |
| An effort to design a traffic safety information management platform based on cloud computing, data warehousing and
data mining, which serves as an outstanding feed for communication between traffic authority, workers and the public
[23]. It aims to take complete benefit of road safety information database in effort to simplify, critic and envisage
different safety information such as traffic accident black-spots, vehicle types liable to severe accidents and climate
based on previously occurred accidents on one hand and collect, deal with and divide complete traffic circumstances
without delay such as road condition, potential risks, accidents, disasters and weather on the other hand. The key
problem requests tackling. |
| There are three key technologies: |
| • Cloud storage of huge traffic data |
| • precise analysis and proposal of massive traffic information |
| • Cloud terminal technology |
| Support different mobile terminal including mobile phones, ipad and personalized cloud terminal in various interactive
forms so that users can use cloud services anytime and anywhere and distribute traffic safety information. |
Architecture design of intelligent traffic safety information platform based on cloud computing |
| Based on the responsiveness and understanding of above problem, attempt to design the following system structure.
Figure 2. Briefly explained as follows. Those bunch data gathered by each branch system grant support for isolating
users in the field of traffic safety. In most cases driver’s casual behaviors often result in uneven pattern, which is further
suitable for cluster analysis. |
| Due to the large scale of user data, we choose to take up Hadoop for parallel computing and distributed storage.
Moreover, to advance the consumption rate of resources we have realized the virtualization of server layer, built traffic information security awareness based on all kinds of data mining algorithm with Mahout and blueprint mobile terminal
access for solving all kinds of cloud terminal access in the application layer. |
Algorithm usage case |
| There is no extensively accepted effective algorithm on report of incomplete statistical data so far. It is an attempt to
confine influencing factors of road safety in a grey realm by adopting clustering algorithm [24] and estimate overall
level of road traffic safety based on information screening, processing, annex and extension. |
| The procedures will be as follows: |
| • Format conversion: Data format have to be transformed into the input format that can be handled by clustering
algorithm. What clustering algorithm can deal directly with the format is sequence file in Mahout. So we
should write a Class which is used to convert the format into sequence file file: |
| InputMapper and a Map function to implement InputMapper. |
| Map function is defined as follows: |
| “Public void Map (LongWritable key, |
| Text values, |
| OutputCollector<Text, VectorWritable>Output, |
| Reporter reporter)” |
| • Invoke Apache Mahouts |
| • Invoke parallel clustering algorithm of Mahout, which contains a couple of clustering algorithm. K - means
clustering algorithm is a case in point. |
| • Obtain the clustering results of HDFS |
| • Analyze the clustering results, namely obtain the results directly in HDFS by Hive and extract them to local
client for analysis. |
CONCLUSION |
| It is not so simple to begin from graze and thus call for additional analysis and exploration. Its purpose in the traffic
field also depends on the expansion of related areas such as IT and logistics. Due to its potential theoretical and
practical values in the field of traffic management, cloud computing has aroused wide concern among researchers. The
expansion and application of cloud computing in different vital industries can also make possible its progress in traffic
field. |
| |
Tables at a glance |
 |
| Table 1 |
|
| |
Figures at a glance |
 |
 |
| Figure 1 |
Figure 2 |
|
| |
References |
- Shi J., Li, X. (2011). Research on Traffic Information Cloud Computing and Its Application. Journal of Transportation Systems Engineering and Information Technology,(01):179-184.
- Zeng, K., Yan, J. (2011). Cloud Computing and Its Application in Intelligent Transportation. Modern Science &Technology of Telecommunication,(05):45-51.
- Debasish Jana and Debasis Bandyopadhyay, "Management of Identity and Credentials in Mobile Cloud Environment", Proceedings of the 2013 International Conference on Advanced Computer Science and Information Systems (ICACSIS 2013), September 28-29, 2013, Bali, Indonesia
- Martin Pirker, Daniel Slamanig, and Johannes Winter. 2012. Practical privacy preserving cloud resource-payment for constrained clients. In Proceedings of the 12th international conference on Privacy Enhancing Technologies (PETS'12), Simone Fischer-Hübner and Matthew Wright (Eds.). Springer-Verlag, Berlin, Heidelberg, 201-220.
- Mahadev Satyanarayanan. 2013. Cloudlets: at the leading edge of cloudmobile convergence. In Proceedings of the 9th international ACM Sigsoft conference on Quality of software architectures (QoSA '13). ACM, New York, NY, USA, 1-2.
- M. Abuelela and S. Olariu, “Taking vanet to the clouds,” Proceedings of The 8th International Conference on Advances in Mobile Computing and Multimedia MoMM 2010, pp. 8–10, 2010.
- N. Santos, K. P. Gummadi, and R. Rodrigues, “Towards trusted cloud computing,” in Proceedings of HotCloud, June 2009.
- F. J. Krautheim, “Private virtual infrastructure for cloud computing,” 2009 conference on Hot topics in cloud computing, pp. 1–5, 2009.
- M. Abuelela and S. Olariu, “Taking vanet to the clouds,” Proceedings of The 8th International Conference on Advances in Mobile Computing and Multimedia MoMM 2010, pp. 8–10, 2010.
- M. Eltoweissy, S. Olariu, and M. Younis, “Towards autonomous vehicular clouds,” in Proceedings of AdHocNets’2010, Victoria, BC, Canada, August 2010.
- S. Olariu, I. Khalil, and M. Abuelela, “Taking vanet to the clouds,” International Journal of Pervasive Computing and Communication, vol. 7, no. 1, pp. 7–21, 2011.
- T. Ristenpart, E. Tromer, H. Shacham, and S. Savage, “Hey, you, get off of my cloud: exploring information leakage in thirdparty compute clouds,” in Proceedings of the 16th ACM conference on Computer and communications security, ser. CCS ’09, 2009, pp. 199–212.
- Thomas Ristenpart, Eran Tromer, Hovav Shacham, and Stefan Savage. 2009. Hey, you, get off of my cloud: exploring information leakage in third-party compute clouds. In Proceedings of the 16th ACM conference on Computer and communications security (CCS '09). ACM, New York, NY, USA, 199-212.
- Rackspacecloud. http://www.rackspace.com/cloud,Accessed April 10th, 2012.
- Google app engine. https://cloud.google.com/products/app-engineappengine-search Accessed November 15th, 2011.
- Google apps for business. https://www.google.com/enterprise/apps/business/products.html Accessed April 10th, 2012.
- Salesforce. http://www.zdnet.com/why-salesforce-integrator-rework-is-recasting-itself-as-a-cloud-broker-7000027448/ Accessed April 10th, 2012.
- Amazon elastic compute cloud (ec2), https://www.google.co.in/url. Accessed December 10th, 2011.
- Microsoft azure. http://www.localmoxie.com/web.php Accessed April 10th, 2012. [Online].
- Basics About Cloud Computing, Grace Lewis, September 2010.
- Zheng, X. (2012). Application of Cloud Computing in the Intelligent Transportation System in the Future. Guangxi Journal of Light Industry,(03):88-89.
- White, T. (2011). Hadoop: The Definitive Guide. : 30-32 About Mahout https://cwiki.apache.org/MAHOUT/quickstart.html
- Gray Evaluation Method Based on the Relative Merits of the Clustering of Road Traffic Safety. Road Traffic & Safety. 2006(11)
- Jaworski, P., Edwards, T., Moore, J., Burnham, K. (2011). Cloud Computing Concept for Intelligent Transportation Systems. 2011 14th International IEEE Conference on Intelligent Transportation Systems. 2011: 931-936
|
Menu