Recent advancements of biometrics identity verification are growing rapidly in this vastly interconnected techno-savvy society. In this information age, protection of valuable contents from the unauthorised intruders or illegal entry to high security zones has made these biometric systems crucial mechanism towards establishing a robust identity verification system. The thrust for reliable authentication methodologies are increasing due to security consciousness of people and also for growing advancement of civilian infrastructures by means of networking, communication, E-Governance, IT knowledge-based civic environment, etc. In the last two decades, a large number of computational intelligence (CI) based and non-linear synchronization based approaches have been thoroughly investigated in biometric authentication in terms of automatic feature detection, feature matching and association of adaptive parameters to the system. Although, it has been felt that the robust and invariant ways are necessary to process the system development from one biometric application to another. However, some incapable and negative constraints have made these biometric systems lack of inconvenience to a large group of end users. To cope up with these incapable factors in biometric systems successfully, Scale Invariant Feature Transform (SIFT) operator has been thoroughly investigated and proved to be invariant to image rotation, scaling, partly illumination changes, biometric authentication towards efficient identity verification.

After several decades of effort, significant progress has been made in the area of speech recognition technologies, and various speech-based applications have been developed. However, current speech recognition systems still generate erroneous output, which hinders the wide adoption of speech applications. Given that the goal of error-free output can not be realized in near future, mechanisms for automatically detecting and even correcting speech recognition errors may prove useful for amending imperfect speech recognition systems. This dissertation research focuses on the automatic detection of speech recognition errors for monologue applications, and in particular, dictation applications. Due to computational complexity and efficiency concerns, limited linguistic information is embedded in speech recognition systems. Furthermore, when identifying speech recognition errors, humans always apply linguistic knowledge to complete the task. This dissertation therefore investigates the effect of linguistic information on automatic error detection by applying two levels of linguistic analysis, specifically syntactic analysis and semantic analysis, to the post processing of speech recognition output. Experiments are conducted on two dictation corpora which differ in both topic and style (daily office communication by students and Wall Street Journal news by journalists). To catch grammatical abnormalities possibly caused by speech recognition errors, two sets of syntactic features, linkage information and word associations based on syntactic dependency, are extracted for each word from the output of two lexicalized robust syntactic parsers respectively. Confidence measures, which combine features using Support Vector Machines, are used to detect speech recognition errors. A confidence measure that combines syntactic features with non-linguistic features yields consistent performance improvement in one or more aspects over those obtained by using non-linguistic features alone. Semantic abnormalities possibly caused by speech recognition errors are caught by the analysis of semantic relatedness of a word to its context. Two different methods are used to integrate semantic analysis with syntactic analysis. One approach addresses the problem by extracting features for each word from its relations to other words. To this end, various WordNet-based measures and different context lengths are examined. The addition of semantic features in confidence measures can further yield small but consistent improvement in error detection performance. The other approach applies lexical cohesion analysis by taking both reiteration and collocation relationships into consideration and by augmenting words with probability predicted from syntactic analysis. Two WordNet-based measures and one measure based on Latent Semantic Analysis are used to instantiate lexical cohesion relationships. Additionally, various word probability thresholds and cosine similarity thresholds are examined. The incorporation of lexical cohesion analysis is superior to the use of syntactic analysis alone. In summary, the use of linguistic information as described, including syntactic and semantic information, can provide positive impact on automatic detection of speech recognition errors.

Much work has been done on perfecting Service-Oriented Architectures (SOAs) in the case of connected networks and networks which have full-time access to centralized service directories. Networks which have sparse connectivity, mobile nodes, and limited bandwidth cannot use this model for a SOA. Many of today and tomorrow’s tactical networks will have these limitations. We have designed and built a SOA which combines the advantages of swarming technology and active metadata, extending SOA and service invocation capability to these tactical edge networks. Autonomous mobile nodes within our swarming architecture also have the capability to reconfigure the edge network topology to optimize service response time, while at the same time completing complementary tasks such as area search. Our architecture is modeled on the successful Representational State Transfer (REST) architecture for Web Services. Metadata and service invocation are therefore very lightweight and are capable of easily being transported through the limited-capability swarming network. Service responses can either be transmitted via direct link, if one exists, or can be transported via mobile swarming nodes for network delay-tolerance. We present the results of testing our architecture in a ChemBio scenario during an officially-sponsored autonomy experiment.

In a highly dynamic environment, knowledge of currently available and relevant resources is essential to the development of comprehensive situational awareness and assessment of threat level or status. We present an agent-based framework, the Active Metadata Framework, which supports the intelligent distribution of resource information among diverse assets, in a manner which supports focused awareness of resources and provides some measure of network disruption tolerance. This framework supports the concept of power to the edge by rapidly moving key data products directly to front-line warfighters. This accelerated movements of information improves war-fighter effectiveness by shortening the observeorient-decide-act (OODA) loop. This framework is demonstrated within a simulated complex military engagement in which assets are required to identify information sources (such as other similar assets, sensors, or databases), and utilize the data they provide to infer information about their situation and required actions. Inference in this framework is performed using Markov Logic Networks (MLNs) with temporal extensions. Preliminary results are presented which demonstrate the successful, ad-hoc creation of networks of assets using this framework, the effective distribution of data through this dynamic network of assets, and the ability to infer information of value from the available data.

An essential step in fielding a timely and effective response to events of global importance is the ability to rapidly identify and integrate a crisis action team. This group should consist of exactly those individuals best qualified to manage the situation. Often, the organization of such a team follows identifiable patterns. Thus, it is important to rapidly identify the type of team, or pattern, required, and to identify the individuals that meet the requirements specified by this pattern. This is a challenging task, as information about people is often distributed across multiple locations, inconsistent or out-of-date, and phrased in the language of different domains. We present a framework that facilitates the rapid integration of teams by identifying scenario-based patterns, and using agent-based search across enterprise boundaries to identify people and assist in their assignment. George Abitante, Michael Cramer, Steve Forsythe, Timothy Frey, Anil John, and Kim Richeson contributed to the concepts and technical discussions presented in this paper. The authors appreciate their inputs and valuable insights.

There are many scenarios where there is a need for (semi) automated methods and tools to identify, characterize and exploit information resources, especially those that may have been discovered through obscure means. These information resources are retrieved from environments where there is little to no prior knowledge of the information sources, and from environments where there are unavailable models and uncooperative modelers. The key to this capability is developing techniques for crafting an understanding of the content and context of an information resource, and ultimately reconstructing the meaning and intent of the resource. Our approach to inferring meaning and intent is to gather the implicit semantics available in data source schemas, to build associations between the contents of the data source and the semantics described and defined in ontologies, and to glean additional semantic clues captured from an analysis of a set of queries submitted to the data source.

Abstract Anessential step in fielding a timely and,effective response to events ,of global importance is the ability to rapidly identify and integrate a crisis action team. This group should consist of exactly those individuals best qualified to manage the situation. Often, the organization of such a team follows identifiable patterns. Thus, it is important to rapidly identify the type of team, or pattern, required, and to identify the individuals that meet

Tcl/Tk is an attractive language for the design of intelligent agents because it allows the quick construction of prototypes and user interfaces; new scripts can easily be bound at runtime to respond to events; and execution state is encapsulated by the interpreter, which helps in agent migration. However, a system of intelligent agents must share a common language for communicating requests and knowledge. We have integrated KQML (Knowledge Query Manipulation Language), one such standard language, into Tcl/Tk. The resulting system, called TKQML, provides several benefits to those building intelligent agent systems. First, TKQML allows easy integration of existing tools which have Tcl/Tk interfaces with an agent system by using TO to move information between KQML and the application. Second, TKQML is an excellent language with which to build agents, allowing on-the-fly specification of message handlers and construction of graphical interfaces. This paper describes the implementation of TKQML, and discusses its use in our intelligent agent system for information retrieval.

Swoogle is a crawler-based indexing and retrieval system for the Semantic Web. It extracts metadata for each discovered document, and computes relations between documents. Discovered documents are also indexed by an information retrieval system which can use either character N-Gram or URIrefs as keywords to find relevant documents and to compute the similarity among a set of documents. One of the interesting properties we compute is ontology rank, a measure of the importance of a Semantic Web document.

Swoogle is a crawler-based indexing and retrieval system for the Semantic Web. It extracts metadata for each discovered document, and computes relations between documents. Discovered documents are also indexed by an information retrieval system which can use either character N-Gram or URIrefs as keywords to find relevant documents and to compute the similarity among a set of documents. One of the interesting properties we compute is ontology rank, a measure of the importance of a Semantic Web document.

For initial ranked retrieval, we continue to use a statistical language model to compute query/document similarity values. Hiemstra and de Vries [3] describe such a linguistically motivated probabilistic model and explain how it relates to both the Boolean and vector space models. The model has also been cast as a rudimentary Hidden Markov Model [4]. Although the model does not explicitly incorporate inverse document frequency, it does favor documents that contain more of the rare query terms. The similarity measure can be computed as

We present the results of UMBC’s participation in the Web and Nov- elty tracks. We explored various heuristics-based link analysis approaches to the Topic Distillation task. For the novelty task we tried several methods for exploit- ing semantic information of sentences based on the SVD technique. We used SVD to expand the query and to filter redundant sentences. We also used a clus- tering algorithm that is also based on SVD.

Darpa Agent Markup Language (DAML) [7] is the newest effort for Semantic Web [5]. It can be used to create ontologies and markup information resource like web pages. The information resource can be read by human and understood by agent programs. We believed DAML could be used to markup agent communication content and promote knowledge sharing and exchanging between agents. This paper also suggested an alternative model to connect web and agent together. We defined the necessary ontologies for agent communication in DAML language and described the agent communication scenario occurred in the ITTalks Project.

Effective use of the vast quantity of available information and services on the Internet will require multi-agent systems to be tightly integrated with existing web infrastructure. This however will be impossible unless the information on the web is presented in a semantic language, such as the DARPA Agent Markup Language (DAML), which is one aim of the “Semantic Web”. As part of our exploration of Semantic Web technology, and DAML in particular, we have constructed ITTALKS, a web-based system for automatic and intelligent notification of information technology talks. In this paper, we describe the ITTALKS system, and discuss the numerous ways in which the use of Semantic Web concepts and DAML extend its ability to provide an intelligent online service to both the human community and, more interestingly, the agents assisting them.

We describe CARROT II (C2), an agent-based architecture for distributed information retrieval and document collection management. C2 can consist of an arbitrary number of agents, distributed across a variety of platforms and locations. C2 agents provide search services over local document collections or information sources. They advertise content-derived metadata that describes their local document store. This metadata is sent to other C2 agents which agree to act as brokers for that collection, and every agent in the system has the ability to serve as such a broker. A query can be sent to any C2 agent, which can decide to answer the query itself from its local collection, or to send the query on to other agents whose metadata indicate that they would be able to answer the query, or send the query on further. Search results from multiple agents are merged and returned to the user. C2 differs from similar systems in that metadata takes the form of an automatically generated, unstructured feature vector, and that any agent in the system can act as a broker, so there is no centralized control. We present experimental results of retrieval performance and effectiveness in a distributed environment.

Semantic Web markup languages will improve the automated gathering and processing of information and help integrate multiagent systems with the existing information infrastructure. The authors, describe their ITtalks system and discuss how Semantic Web concepts and DAML+OIL extend its ability to provide an intelligent online service.

The Semantic Web is a vision to simplify and improve knowledge reuse on the Web. It is all set to alter the way humans benefit from the web from active interaction to somewhat passive utilization through the proliferation of software agents and in particular personal assistants that can better function and thrive on the Semantic Web than the conventional web. Agents can parse, understand and reason about information available on Semantic Web pages in an attempt to use it to meet users’ needs. Such personal assistants will be driven by rules , axioms and the internal model or profile that the agents have inside them for the user. An intrinsic and important pre-requisite for a personal assistant or rather any agent is to manipulate information available on the Semantic Web in the form of ontologies, axioms, and rules written in various semantic markup languages. In this paper, a model architecture for such a personal assistant dealing with real-world semantic markup is described. The agent reasons with semantic markup written in DAML+OIL, using the Java Expert System Shell (JESS) as the reasoning engine. This software assistant views information providers on the Semantic Web as recommender agents that have a limited view of the user’s preferences and provides a improved notion of personalization by collaborating with peer personal assistants (what are referred to as buddy agents) within communities that the user has identified as trusted parties to exchange information with. Collaboration is achieved through simple solicitation and recommendation of information with these buddy agents.