Visualization for levels of animal diseases by integrating OLAP and GIS

Shared by: ijcsiseditor

Categories

Technology > Computers & Internet

Stats

views:: 86
posted:: 8/19/2012
language:: English
pages:: 7

Public Domain

Document Sample

(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 10, No. 7, July 2012

Visualization for Levels of Animals Diseases by Integrating
OLAP and GIS
Hesham Ahmed Hassan Hazem El-Bakry Hamada Gaber Abd Allah
Faculty of Computer and Information, Faculty of Computer and Information, Faculty of Computer and Information,
Cairo University Sciences, Mansoura University Sciences, Mansoura University
Giza, Egypt Mansoura, Egypt Mansoura, Egypt

Abstract increasing gap between dairy products produced
Animal diseases have constituted a major problem in many domestically and the amount consumed. The gap between
developing and developed countries. There are different domestic animal production and consumption has been
limitations for the existing computer systems to meet the estimated at an average of 17 per cent for red meat and 19
required information and analytical capabilities for a better per cent for milk. This gap has been continuously widening
decision in the Egyptian animal production domain. This paper
presents an approach for helping policy/decision makers to
over recent years and consequently dependence on food
improve animal production in Egypt. The paper integrates Online imports has been increasing [1]. In 2000 population of
Analytical Processing (OLAP), Geographical Information dairy animals in Egypt was about 6.7 million heads of
System (GIS), Spatial Analysis functions and Multicriteria cattle and buffaloes contributing about 30% of the total
Decision Analysis (MCDA) capabilities to develop a Spatial value of agricultural production. [2].
Decision Support System (SDSS). The main aim of this study is
to generate a composite map for decision makers by using some The agricultural domain in Egypt plays a crucial role in the
effective factors affect animal production in Egypt. We visualize national economy as it represents 20% of GDP and
and analyze different factors such as "Diseases", "Climate", "Soil employs nearly 30% of the working population. Also, the
Pollution", "Veterinary care" and "Economical factors" which
affect the animal production in Egypt. The paper takes in
feeding adequately a population growing at an annual rate
consideration influence of each factor because importance and of 1.8%, with limited water resources and land, is
influence of each factor differs according policy/decision makers considered as the most important challenge for policy
point of view. makers in Egypt. In addition, the national food security has
been noted to be the main goal to achieve a real
Keywords: Geographical Information System (GIS), development and to meet rising of the Egyptian population
Multicriteria Decision Analysis (MCDA), Online Analytical that expected to be more than 100 million by the year 2030.
Processing (OLAP), Spatial Analysis and Spatial Decision The policy/decision makers’ strategy for animal production
Support System (SDSS). in Egypt, up to year 2037, aims to reduce the milk
production gap to be less than 10% [3].
1. Introduction Geographical Information System (GIS) links a location
and attribute information and enables a person to visualize
Food crises in less-developed countries have been noted to patterns, relationships, and trends. This process gives an
be the main obstacle to economic development. Moreover, entirely new perspective to data analysis that cannot be
feeding adequately a population growing at an annual rate easily seen in a table or list format or on a paper map.
of 2.1 %, with limited land and water resources, is Exploring data using GIS turns data into information into
considered the most important challenge for Egypt. The knowledge. There are two ways that the layers of location
population of 74 million is expected to rise to 90 million by can be visualized on a map: Raster layers are organized in a
the year 2017. The high population growth rate is a major grid of identically sized cells. The cells have a uniform
constraint for sustainable development in Egypt. In Egypt length and width (square shaped) and are called “pixels.”
the population dynamics tells interesting situation: dairy Vector layers are represented as points, lines, or polygons.
cattle -5.3%, buffaloes +12.1%, beef cattle +50.0%, sheep A vector layer cannot mix types together. One layer cannot
+29.9%, goats +32.8%, while people numbers increased have both points and polygons. The layer would have to be
more than 18%. Nevertheless, there is a shortage of protein split into two separate layers; one for points and one for
and calcium from animal sources produced in Egypt in polygons. Vector data is used when the features have
comparison to nutritional requirements, and there is an specific locations and boundaries and the attribute data is

44 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 10, No. 7, July 2012
(

uniform throughout the individual features. Examples of What really makes the difference between a SDSS (Spatial
vector layers include bus stops (point), roads (line), and Decision Support System) and a traditional DSS (Decision
counties (polygon). Support System) is the particular nature of the geographic
data considered in different spatial problems. In addition,
Transactional systems are not designed to support the
traditional DSSs are devoted almost only to solve
decisional processes, new types of systems have been
structured and simple problems which make them non
developed to specifically fulfill decisional needs; they are
practicable for complex spatial problems [7]. SDSS
called “Analytical Systems” and are known on the market
requires the addition of a range of specific techniques and
as “Business Intelligence” (BI) solutions. In the BI world,
functionalities used especially to manage spatial data, to
data warehouses are based on data structures called
conventional DSSs. These additional capacities enable the
“multidimensional”. The term “multidimensional” was
SDSS to [6];
coined in the mid-1980s by the community of computer
scientists who were involved in the extraction of • acquire and manage the spatial data,
meaningful information from very large statistical
• represent the structure of geographical objects and
databases (ex. national census). The most widely used BI
their spatial relations,
solutions are OLAP (On-Line Analytical Processing)
systems, which provide a unique capability to interactively • diffuse the results of the user queries and SDSS
explore the data warehouse. OLAP technology is based on analysis according to different spatial forms
the multidimensional database approach, which introduces including maps, graphs, etc., and to
concepts that differ from the concepts found in the
• Perform an effective spatial analysis by the use of
transactional database approach. The key multidimensional
specific techniques.
concepts include: dimensions, members, measures, facts
and data cubes [4]. A cube is a multidimensional structure Multi-criteria decision making (MCDM) refers to making
that contains dimensions and measures. Dimensions define decisions for alternatives in the presence of multiple and
the structure of the cube, and measures provide the conflicting criteria. A main contribution area of MCDM is
numerical values of interest to the end user. making preference decision (e.g., evaluation, prioritization,
selection) over the available alternatives such as a set of
OLAP systems are expected to [5]:
products that are characterized by multiple, usually
• Provide ad hoc access.
conflicting attributes [8].
• Support the complex analysis requirements of
decision-makers.
• Analyze the data from a number of different 2. Problem Formulation
perspectives (business dimensions).
The Central Laboratory for Agriculture Expert Systems
• Support complex analyses against large input (CLAES) in Egypt hosts the data base of Bovine
(atomic-level) datasets. Information System (BOVIS) project that has more than 2
million records represented in 52 tables. In this paper we
In order to improve the efficiency and response time of the use El Sharkeya Governorate as case study. [2]Tables
Data Warehouse, the preferred structure is the Star Schema. related to cow or buffalo sex, major disease categories,
Star Schemas a database structure in which data is various diseases and disorders that affect them, the breeds,
maintained in a single fact table located at the center of the the governorate, directorates and the veterinary units they
schema with additional dimension data stored in are affiliated to were classified for mining. As data
dimensional tables, with all hierarchies collapsed. production and collection is escalating.
Decision makers have turned to analysts and analytical The purpose of this paper is to do the following:
modeling techniques to enhance their decision making 1. Building OLAP (Online Analytical
capabilities. Spatial decision support systems (SDSS) are Processing) system instead of TPS
explicitly designed to support a decision research process (Transaction Processing System).
for complex spatial problems. SDSS provide a framework 2. Visualizing OLAP output dimensions using
for integrating database management systems with Geographical Information System (GIS).
analytical models, graphical display and tabular reporting 3. Using GIS Spatial Analysis capabilities.
capabilities, and the export knowledge of decision makers. 4. Building Spatial Multiple Criteria Decision
Such systems can be viewed as spatial analogues of Analysis for different factors diseases,
decision support systems (DSS) developed in operational Climate, Soil pollution and Economical
research and management science to address business factor see Fig (1).
problems [6].

45 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 10, No. 7, July 2012

Fig 3. a: Web-Based OLAP Dundas Visualization (Grid)

Fig 3. b: Web-Based OLAP Dundas Visualization (Bar Charts)
Fig 1. General Workflow of Multicriteria Evaluation (MCE)
Web Based Dundas tool allows users to select dynamic
cubes and determine measures and dimensions. Users can
3. Proposed Method choose any cube such as "card_animal", "death","
disorder", "pregnancy", "slaughters", "vaccine" …etc (see
3.1 Building OLAP Database Fig 2). Also users can specify way of display data either
Grid or Bar Charts.
There is an existing OLAP database for BOVIS
project build by CLAES team. OLAP see BOVIS from
different dimensions such as animal count, deaths, 3.2 Visualizing OLAP Output Dimensions
disorders/disease, and pregnancy …etc Fig (2).
In these step we use GIS engine to visualize OLAP
dimensions by preparing data in ArcCatalog GIS using
feature classes and relationship class for El Sharkeya
governorate.
Feature classes are homogeneous collections of common
features, each having the same spatial representation, such
as points, lines, or polygons, and a common set of attribute
columns see Fig (4).

Fig 2. BOVIS OLAP Cubes and Dimensions

An OLAP system is built especially to navigate within
multidimensional cubes, i.e., to go from one fact to
another in an interactive manner and to obtain fast
responses. We visualize OLAP multidimensional cubes
using Web based Dundas OLAP services and ASP.Net see
Fig (3).

Fig 4. Feature Class Properties

Three layers namely: "Veterinary Units", "Climate" and
"Economical Standard of Living" are represented as
Polygon feature class. Each disease is represented by

46 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 10, No. 7, July 2012
(

Geodatabase table. Relationships classes in the 3.4 Drive New Data Layers (Raster)
Geodatabase manage the associations between objects in
one class (feature class or table) and objects in another [5]. Prepare and unify layers format to be Raster data. There are
Objects at either end of the relationship can be features several ways to think about converting raster data in
with geometry or records in a table. ArcGIS. You may want to convert non raster data into
raster data or vice versa, such as converting a polygon into
3.3 Editing Layers using ArcMap. a raster. "Diseases", "Economical", "Soil Pollution" and
"Climate" layers are converted from Polygon to Raster.
We use editing tools of ArcMap 10 to edit "Veterinary
Units" layer on the map see Fig (5.a). All diseases layers
joined with "Veterinary Units" layer see Fig (5.b).

Fig 5. a: El Sharkeya Governorate Map with Veterinary
Units Fig 6. Convert Polygon to Raster

3.5 Reclassify Data

Reclassify data to values range from 1 to 9, all data
reclassified to give weights. 9 is the most suitable value for
animal production and 1 is the least.

Fig 7. Reclassify Raster Data Layers
Fig 5. b: El Sharkeya Governorate Map with Diseases Count in Each
Veterinary Unit
3.6 Weight and Combine Layers
Overlays several raster using a common measurement
scale and weights each according to its importance. Seven
diseases layers weighted using weighted overlay see Fig

47 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 10, No. 7, July 2012

(8). Output layer of weighted diseases weighted with • Each input raster is weighted according to its
"Economical", "Soil Pollution" and "Climate" layers see importance or its percent influence. The weight is
Fig (9). a relative percentage, and the sum of the percent
influence weights must equal 100.
Overlays several raster using a common measurement scale
and weights each according to its importance. Seven • Changing the evaluation scales or the percentage
diseases layers weighted using weighted overlay see Fig influences can change the results of the weighted
(8). Output layer of weighted diseases weighted with overlay analysis.
"Economical", "Soil Pollution" and "Climate" layers see
Fig (9).

Fig 8.a: Weighted Overlay Diseases

Fig 10. Spatial Multiple-Criteria Workflow

4. Results
Weighted overlay spatial analysis of diseases results
indicate the following see Fig (11):
Fig 8.b: Weighted Overlay Influence
• Worst veterinary unit in EL Sharkeya governorate
is Kofor Negm unit. This unit contains the highest
diseases frequency.
• Best veterinary units are El Qeniat, El Zenkalon,
Belbess, El Azezia and El Ketawia.
There are different units in middle diseases frequency such
as El Sanafen, Mashtol El Soq and El Balashon.

Fig 9. Weighted Overlay for All Factors with Different
Influence

• All input raster must be integer. A floating-point
raster must first be converted to an integer raster
before it can be used in Weighted Overlay.
• Each value class in an input raster is assigned a
new value based on an evaluation scale. These
new values are reclassifications of the original
input raster values. A restricted value is used for Fig 11. Diseases Weighted Overlay Results
areas you want to exclude from the analysis.

48 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 10, No. 7, July 2012
(

Diseases are an important factor in animal production. For their influence on the decision making. For instance, we
instance, we supposed the following: supposed the following:
• The weighted diseases output layer influence is • The weighted diseases output layer influence is
50%. 50%.
• Economical factor influence represents 18%. • Economical factor influence represents 18%.
• Soil Pollution and Climate factors influence • Soil Pollution and Climate factors influence
represent 16% for each factor. represent 16% for each factor.
Influence of each factor can be changed according its Anyway the influence of each factor can be changed
importance. The result of weighted overlay for factors according its importance at any time. The result of the
affects animal production in Egypt represented in Fig (12). weighted overlay for factors that affects animal production
The value 3 represents the worst places for animal in Egypt is represented in Fig (12). As shown in this figure
production in EL Sharkeya governorate and the value 8 the value 3 represents the worst places for animal
represents the best places as in Fig (12). production in EL Sharkeya governorate and the value 8
represents the best places.

Acknowledgment

The authors wish to acknowledge the Central Laboratory
for Agriculture Expert Systems (CLAES) in Egypt and
ESRI Support Center.

References
[1] S. Gamal and H.Moussa, "Food Security in Egypt Under
Economic Liberalization Policies and WTO Agreement",
International Conference Agricultural policy reform and the
WTO: where are we heading? Italy, 2007.
[2] El Fangary, L.M.; , "Mining Data of Buffalo and Cow
Production in Egypt," Frontier of Computer Science and
Technology, 2009. FCST '09. Fourth International
Fig 12. Weighted Overlay for All Factors Affect Animal Production
Conference on , vol., no., pp.382-387, 17-19 Dec. 2009
in Egypt.
doi:10.1109/FCST.2009.27
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnum
ber=5392891&isnumber=5392815.
5. Conclusion [3] Omran, A.; Khorshid, M.; Saleh, M.; , "Intelligent decision
support system for the Egyptian food security," Intelligent
This paper presents an approach for helping Systems Design and Applications (ISDA), 2010 10th
policy/decision makers to improve animal production in International Conference on , vol., no., pp.557-562, Nov. 29
Egypt. We visualize and analyze different factors such as 2010-Dec. 1 2010
"Diseases", "Climate", "Soil Pollution", "Veterinary care" doi: 10.1109/ISDA.2010.5687207
and "Economical factors" which affect the animal URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnum
production in Egypt. The paper takes in consideration ber=5687207&isnumber=5687016.
influence of each factor because importance and influence [4] Rivest, S., Bédard, Y., Proulx, M.-J., Nadeau, M., Hubert, F.,
& Pastor, J. (2005). "SOLAP technology: Merging business
of each factor differs according policy/decision makers
intelligence with geospatial technology for interactive spatio-
point of view. In this research we aim to present the best temporal exploration and analysis of data". ISPRS Journal of
way to visualize animal diseases and find the best and Photogrammetry and Remote Sensing, 60(1), 17-33.
worst places in EL Sharkeya Governorate for animal doi:10.1016/j.isprsjprs.2005.10.002
production. We use weighted overlay spatial analysis to [5] Ahsan Abdullah “Analysis of mealybug incidence on the
indicate that the worst veterinary unit in EL Sharkeya cotton crop using ADSS-OLAP (Online Analytical
Governorate is Kofor Negm unit. This unit contains the Processing) tool”, Computers and Electronics in
highest diseases frequency and with weight equal 3, where Agriculture,2009.
as the best veterinary units are El Qeniat, El Zenkalon, [6] P.J. Densham. "Spatial decision support systems". In D.J.
Maguitre, M.F. Goodchild, and D. Rhind, editors,
Belbess, El Azezia and El Ketawia. The later units contain
Geographical Information Systems:Principles and
the lowest diseases frequency and with weight equal to 9. Applications. Longman, London, 1991.
On the other hand we try to present other factors and study

49 http://sites.google.com/site/ijcsis/
ISSN 1947-5500
(IJCSIS) International Journal of Computer Science and Information Security,
Vol. 10, No. 7, July 2012

[7] Maktav, D.; Jurgens, C.; Siegmund, A.; Sunar, F.; Esbah, H.;
Kalkan, K.; Uysal, C.; Mercan, O.Y.; Akar, I.; Thunig, H.;
Wolf, N.; , "Multi-criteria spatial decision support system for
valuation of open spaces for urban planning," Recent
Advances in Space Technologies (RAST), 2011 5th
International Conference on , vol., no., pp.160-163, 9-11
June 2011
doi: 10.1109/RAST.2011.5966812
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnum
ber=5966812&isnumber=5966798.
[8] Yoon K, Hwang C, "Multiple attribute decision making – An
introduction", SAGE Publications, Inc., Thousand Oaks,
1995.

Hesham A. Hassan is an Egyptian researcher born in Cairo in 1953.
Hesham's educational background is as follows: B.Sc in Agriculture,
Cairo University, Egypt in 1975. Postgraduate diploma in computer
science, from ISSR, Cairo University, Egypt in 1984. M.Sc in computer
science, from ISSR, Cairo university, Egypt, in 1989. Ph.D in computer
science from ISSR, Cairo University (dual supervision Sweden/Egypt) in
1995. He is now a PROFESSOR and HEAD of computer science
department at the faculty of computers and Information, Cairo University.
He is also IT Consultant at Central Laboratory of Agricultural Expert
System, National Agricultural Research Center. He has published over
than 51 research papers in international journals, and conference
proceedings. He has served member of steering committees and program
committees of several national conferences. Hesham has supervised over
27 PhD and M. Sc theses. Prof. Hesham interests are Knowledge
modeling, sharing and reuse, intelligent information retrieval, Intelligent
Tutoring systems, Software Engineering. Cloud Computing and Service
Oriented Architecture (SOA).

Hazem M. El-Bakry (Mansoura, EGYPT 20-9-1970) received B.Sc.
degree in Electronics Engineering, and M.Sc. in Electrical
Communication Engineering from the Faculty of Engineering, Mansoura
University – Egypt, in 1992 and 1995 respectively. Dr. El-Bakry received
Ph. D degree from University of Aizu - Japan in 2007. Currently, he is
assistant professor at the Faculty of Computer Science and Information
Systems – Mansoura University – Egypt. His research interests include
neural networks, pattern recognition, image processing, biometrics,
cooperative intelligent systems and electronic circuits. In these areas, he
has published many papers in major international journals and refereed
international conferences. According to academic measurements, now the
total number of citations for his publications is 502. The H-index of his
publications is 12 and G-index is 19. Dr. El-Bakry has the United States
Patent No. 20060098887, 2006. Furthermore, he is associate editor for
journal of computer science and network security (IJCSNS) and journal
of convergence in information technology (JCIT). In addition, is a referee
for IEEE Transactions on Signal Processing, Journal of Applied Soft
Computing, the International Journal of Machine Graphics & Vision, the
International Journal of Computer Science and Network Security,
Enformatika Journals, WSEAS Journals and many different international
conferences organized by IEEE. Moreover, he has been awarded the
Japanese Computer & Communication prize in April 2006 and the best
paper prize in two conferences cited by ACM. He has also been awarded
Mansoura university prize for scientific publication in 2010 and 2011. Dr.
El-Bakry has been selected in who Asia 2006 and BIC 100 educators in
Africa 2008.

Hamada Gaber is an Egyptian researcher born in Cairo in 1985. He
received the B.Sc degree in Computer and Information Sciences in 2006
from Assuit University, Egypt. He is currently a Master degree researcher
in Mansoura University, Egypt.

50 http://sites.google.com/site/ijcsis/
ISSN 1947-5500