|
Print Friendly Format
Intelligence and Decision Support through Web Based Integrated Semantic and Geospatial Platforms
 Ashish Sonal Founder and CEO Orkash Technologies Gurgaon India ashish.sonal@orkash.com Ashutosh Sharma Founder member and Smart-Systems Technology Lead Orkash Technologies Gurgaon India Abstract This concept paper articulates the architectural design of a scalable Web application that can integrate geospatial data analysis, visualization and location contextualization capabilities with web services, semantic search and web-mining technologies. This provides for the seamless integration of data sources and different web services or web applications through the single interface. The paper also covers the basics of the architecture from the different perspectives of the implementation approaches. Introduction In today’s world, we have many web applications that provide the seamless integration of the services and data in form of mashups, XML, SOAP etc. Our purpose is to design a different kind of desktop application which would have the web 2.0 compatibility and integration of the web services and applications, built around a GIS platform. We then use this architecture to deliver semantic and geospatial analysis capabilities using web based information sources. The aim of such a solution is to put customizable Internet based intelligence and decision support platforms in the hands of the common user for everyday use, much as we use search engines. Such solutions also have powerful applications across a wide range of areas like search and knowledge management, market and competitive intelligence, web intelligence, advertising, location risk management, disaster management, supply chain and logistics, enterprise security, homeland security and battlefield C3I systems. ORKASH Labs has built a prototype as a technology demonstrator with the objective to test the design parameters and encourage collaboration to achieve the above aims. Compatible with the Web 2.0 technologies and focusing on collaboration and information sharing, the prototype seamlessly integrates the disparate sources of information assimilated from different domains or applications through a single interface. The General Architecture Orkash Labs is developing a web enabled geospatially enhanced solution that automates the process of searching and (semantic and spatial) contextualization of information from a sea of web sources. The application referred to as “Orkash’s Prototype” is a pioneering move toward bringing order into the randomness of the Internet searching and information gathering. The application is Web 2.0 enabled facilitating integration and data interchange through web services and web applications. The key premise of the ORKASH’s Prototype application is to address information and intelligence requirements for catalyzing the decision making process across a wide segment of users. That the discovery and exploitation of geographic information in web pages is quite feasible, and exploitation of such information provides a useful new paradigm for the navigation and retrieval of web information is well established [1]. The integration of web mining and natural language processing (NLP) engines in the backend of the ORKASH’s Prototype provide the ability for identifying events and relationships and their geographic context in unstructured data that resides on the web. The ORKASH’s Prototype can be built around an earth-browser (e.g. NASA World Wind,Google Earth) as a means to architect an integrated platform for deliveringWeb2.0 type analytical web services and data visualization (Raster and Vector data), incorporating semantic search, geospatial contextualization, NLP based location, events and relationships contextualization, and data mining/BI engines/decision support platforms accessing disparate databases (that can be local, LAN or internet based). The other important aspect of the design of this Prototype are the plug-ins for the earth browser based GIS front-end which provide for enhancing the functionalities and features of this application. It, thus, needs not to rely on a single source or resource for redundancy, robustness and scalability of the application. Despite the many limitations in ORKASH’s Prototype system, we find it an extremely powerful means for providing, almost in real time, semantically and spatially contextualized intelligence that is derived from the web. The client front-end, which is a earth-browser based visualisation platform does not attempts to supplant the analytical capabilities of a GIS application, but instead provides the means to the end-users to visualize spatial data and to query web enabled RDBMS and BI applications. The architecture also lends itself (using several existing open source applications) to facilitate the integration of the web services, data, and user created content in the form of mashups, XML, KML, SOAP etc. The following sections explain the architectural design of ORKASH’s Prototype and highlight the capabilities and practical functionalities of such a system in terms of integration of data sources and other application services. The abstract model is shown Figure 1 :  Figure 1: The Abstract Model The above layers represent the different aspects of the application e.g. the ability of the end users to create and share ‘user created content’ using the client interface and connectivity with different data sources for integration at the database level. The application facilitates data mashups at the database level assimilating data from different applications. Detailed Architecture The detailed architecture diagram is shown in Figure 2, covering the design components and modules.  Figure 2: Detailed Architecture Diagram – ORKASH’s Prototype “Base application (Geospatial) Box” is the core component of Orkash’s Prototype which includes the capabilities to visualize the geospatial data using the maps. The maps could be in the form of raster or vector; users have the capabilities to add more vector data and visualize those using the earth-browser GIS platform e.g. a business user can put the locations of their offices with appropriate information and share the information with the other users either using the centralized data services or by transferring the information on its own. The Orkash Apps provides the feature to enhance the functionality of the base application by creating the “Plug-ins” to support more robust applications i.e. scalability of features. By adding the appropriate plugins, users can connect through different data sources and services on web, Internet, and local machines. It thus empowers the users as architecture for adding new functionality and features through plugins to fulfil their needs. This includes the inbuilt features of web enabled earth-browser GIS platform, the client side application, which will lead to the application level integration on the overall platform. “Data Analysis”, as the name suggests, analyzes the data which is either created by the users or by any other application, or crawled from the web. This component includes the Artificial Intelligence based system e.g. Natural Language Processing (NLP), and Data Mining and has the capabilities to generate and query geospatial data, and conduct semantic search based on domain specific ontologies. This feature gives the application the ability to grasp relevance of the information. To make it effective, there are Analysis Engines which analyze the data and information from different sources and integrate those with data based upon some features and profile or domain specific criteria. Users can make use of web services and web applications through the integrated web-browser, or use the semantic analysis engines to examine the information and then integrate it with various user created content or previously analyzed data. The NLP Contextualization Engine (Figure3) is used to contextualize the information or analyzed data with respect to location or georeferencing and extract geospatial context of events and relationships. The NLP engines also analyses the semantic structure of the information and writes it to a database. The user can query the contents semantically and visualize the data through the earth-browser geospatial platform.  Figure3: Architecture – NLP Contextualization Engine (Copyright Orkash Services 2009) The “Web Application & Services” component is devoted to integrate the existing web applications and services such as Mashups, SOAP, XML, RSS, REST, etc. These services are accessible using the URL. It is bidirectional in nature and enabled to retrieve and generate the mashups, RSS, SOAP request/response etc. Importantly, this component provides the platform to run the web applications within the workflow of the Orkash Prototype based on the events captured at runtime. The basic requirement is to develop the various plug-ins for the Prototype and define the different events or data which users want to capture an integrate with the other data sources. The system provides to generate the services using Web 2.0 techniques for interoperability support. This may lead to the interchange of the data from or to different sources of information. It gives the users the power to access the data in real time manner from other applications. The “User Created Content” module facilitates the creation of the content as per the users' requirements. This feature is intended to provide the functionality to share, comment, modify and recreate the contents as a means for creating ‘collective intelligence’. It also facilitates communication, collaboration and sharing of user created content. Using this feature, the most challenging task is to integrate the related content and ‘expressiveness’ of the information. To overcome this problem, the Data Analysis module helps to analyze the data. Another feature is to protect and share the information to the specific users or groups tofacilitate the secure means to share or hide the information. The users can put their proprietary data on their local system in the form of some specific standards like XML, doc, pdf, KML, GML (georeferenced data) etc. and then transfer the data through another secure link or network. On the other side, the transferred data can then be put back to the system by accessing the local system and parse the information using the plug-ins built into the Prototype. There is also the centralized storage system to put the content with the access rights of the users and groups. It also facilitates to integrate other information which resides in server side with the contents. BI Visualization - There are different mechanisms to pass the information to the GIS section of the dashboard e.g. Web services, GML, KML, GeoRSS etc. The dashboard and graphs provide the power of spatial analysis on the vast amounts of data either available in the centralized database or on the local machine. By simply loading the data from the source, and defining the parameters for displaying the data, we can synchronize the results with the help of spatial data on GIS maps. Users can save the analyzed result assimilated through dashboard on their local machine and then distribute to other users for collaborative sharing of work and team-based decision making on the same spatial view. Users also have access to the features to create the spatial data layer dynamically from data analyzed on the dashboard using the facts (cubes) and dimensions. There is a feature which automatically contextualizes the information from the locations, if required. Figure 4 shows the process flow of the content creation, and generation using the different protocols or services through the Orkash’s Prototype interface.  Figure 4: Process flow diagram (User Created Content) Copyright Orkash Services 2009 Challenges and Future Work The application gathers information from multiple domains using data connectors and web services, so the big challenge for the application is to synchronize it over the domain or specific to that domain. Information sources that are inter domain are hard to integrate, and need standardised ontologies which define and categorize the distinct and identical (overlapping) information about the domains. There is no limitation to the data and the data format, so the databases for such an application as described in this paper would become huge over a period of time. There is a need to manage and process such huge databases (centralize) using high performance computing and fast storage devices with efficient communication between those systems. This is again a challenging task. Below are the some of the outlined challenges which we are trying to address in the Orkash’s Prototype:
In a real world application, users would have the freedom to generate any type of content e.g. Text, Images, Videos, Interactive media etc. and add content in the form of their opinions in the form of link, tagging, ratings, social connections either by the use of the web applications or Orkash Prototype. The content can be visible to the end users once they are published by the user by either making it public or giving the permission to read/write to other users or groups. To achieve the sharing of the content between the users or other applications, there are a number of mechanism that exists technically e.g. XML, APIs, Ajax etc. These technologies are designed to implement the standards of the web 2.0 framework. The other aspect of the content sharing is to make available the filtered collaborative content by different dimensions of perspectives. Some of the examples are by the ranking of the content user where he can see the collaborative information based on the aggregated ranking of the series of content. There might be the use of the profile management which is again to contextualize concepts by the different parameters over the domain. To collaborate the user generated content there are several techniques that exists to recombine the content and convert into the desired knowledge or information e.g. Mashups is the best example to collect the data from different sources and collaborate to generate the information which might be useful to others. There are several other techniques that exist for example, remixing, aggregation, embedding etc. Acknowledgement We thank to our reviewers for their valuable suggestions that were used to improve our paper. Bibliography and References
|