Ali I. El-Desouky

AN AGENT DECISION SUPPORT MODULE BASED ON GRANULAR ROUGH MODEL

A multi-agent system (MAS) is a branch of distributed artificial intelligence, composed of a number of distributed and autonomous agents. In a MAS, effective coordination is essential for autonomous agents to achieve their goals. Any decision based on a foundation of knowledge and reasoning can lead agents into successful cooperation; to achieve the necessary degree of flexibility in coordination, an agent must decide when to coordinate and which coordination mechanism to use. The performance of any MAS depends directly on the decisions made by the agents. The agents must therefore be able to make correct decisions. This paper proposes a decision support module in a distributed MAS that is concerned with two main decisions: the decision needed to allocate a task to specific agent/s and the decision needed to select the appropriate coordination mechanism when agents must coordinate with other agent/s to accomplish a specific task. An algorithm for the task allocation decision maker (TADM) and the coordination mechanism selection decision maker (CMSDM) algorithm are proposed that are based on the granular rough model (GRM). Furthermore, a number of experiments were performed to validate the effectiveness of the proposed algorithms; the efficiency of the proposed algorithms is compared with recent works. The preliminary results demonstrate the efficiency of our algorithms.

Ranking distributed uncertain database systems: Discussion and analysis

Large databases with uncertainty became more common in many applications. Ranking queries are essential tools to process these databases and return only the most relevant answers of a query, based on a scoring function. Many approaches were proposed to study and analyze the problem of efficiently answering such ranking queries. Managing distributed uncertain database is also an important issue. In fact ranking queries in such systems are an open challenge. The main objective of this paper is to discuss ranking in distributed uncertain database along with its issued problems. Starting with uncertain data representation, query processing and query types in such systems are discussed along with their challenges and open research area. Top-k query is presented with its properties, as a ranking technique in uncertain data environment, mentioning distributed top-k and distributed ranking problems.

An Automatic Label Extraction Technique for Domain-Specific Hidden Web Crawling (LEHW)

General-purpose search engines (e.g. Google and Yahoo) ignore valuable data that represent 80% of the content on the Web, this portion of Web called hidden Web (HW). Pages in the hidden Web are dynamically generated in response to queries submitted via the search forms. In this paper, a new algorithm for extracting labels from multi-attribute (M-A) search form fields is proposed. A technique for automatic query generation for single-attribute (S-A) search forms is also provided in order to enhance the performance of the overall domain-specific hidden Web crawlers. The innovation of (LEHW) algorithm is its capability to distinguish between (S-A) and (M-A) forms; so that the capability of dealing with both of them, unlike most hidden Web crawlers that ignore either of them. Embedding of the proposed algorithm within the overall framework of the HW crawler is evaluated through experiments using real Web sites. The preliminary results demonstrate the accuracy and precision of the proposed approach for most of the sites considered

Ranking distributed database in tuple-level uncertainty

Ranking in uncertain database environments has gained a great importance recently. Many techniques were introduced to rank uncertain databases and others to rank distributed certain databases. Unfortunately, there are not that much techniques in ranking distributed uncertain databases. This paper proposes a framework that improves ranking processing in the case of uncertain and distributed database. In the proposed framework, new communication and computation-efficient algorithms are investigated for retrieving the top-k tuples from distributed sites. These algorithms are applied in tuple-level uncertainty. The main concern of the proposed algorithms is to reduce the communication rounds utilized and amount of data transmitted while achieving efficient ranking. Experimental results emphasize that both proposed algorithms have a great impact on reducing communication cost. Also, the results clarify that the first algorithm is efficient in the case of a low number of sites while the second achieves better performance in the context of a higher number of sites.

Distributed multi-agent communication system based on dynamic ontology mapping

Communication is the most important feature for meaningful interaction among agents in distributed multi-agent systems. Communication enables agents interaction to achieve their goals. Agent communication languages provide a standard in the protocol and language used in the communication, but cannot provide a standard in ontology, because ontology depends on the subject and concept of the communication. This lack of standardisation is known as interoperability problem. In order to obtain semantic interoperability, agents need to agree on the basis of different ontologies. In this paper, a communication layer is proposed to outline the communication between agents, multiplatform communication system (MPCS) architecture is proposed to provide a highly flexible and scalable system. In addition a dynamic ontology mapping system for agent communication (DOMAC) is proposed based on different mapping approaches.

Prediction of Liver Diseases Based on Machine Learning Technique for Big Data

Liver diseases have produced a big data such as metabolomics analyses, electronic health records, and report including patient medical information, and disorders. However, these data must be analyzed and integrated if they are to produce models about physiological mechanisms of pathogenesis. We use machine learning based on classifier for big datasets in the fields of liver to Predict and therapeutic discovery. A dataset was developed with twenty three attributes that include the records of 7000 patients in which 5295 patients were male and rests were female. Support Vector Machine (SVM), Boosted C5.0, and Naive Bayes (NB), data mining techniques are used with the proposed model for the prediction of liver diseases. The performance of these classifier techniques are evaluated with accuracy, sensitivity, specificity.

A Big Data Framework for Mining Sensor Data Using Hadoop

The data gathered from IOTs is considered of high business value. The IOTs devices sense the natural conditions using sensor network comprised of sensor nodes. Mining of big sensor data for useful knowledge extraction is a very challenging task. Frequent itemsets is one of the most effective mining techniques that find important itemsets from big sensor data. In this paper, a MapReduce Frequent Nodesets-based Boundary POC tree (MR-FNBP) framework is proposed for mining Frequent Nodesets for big sensor data. The MapReduce framework is used to implement MR-FNBP to enhance its performance in highly distributed environments. Additionally, the proposed Boundary (FNBP) creates a Boundary as an early stage to exclude the infrequent itemsets, and this may reduce the overall memory and time usage. Moreover, a number of experiments were performed to evaluate the performance of MR-FNBP framework. The results show high scalability rate and a less time consuming process for MR-FNBP framework over different recent systems.

Discussion and analysis of the distributed uncertain database systems ranking

Large databases with uncertainty became more common in many applications. Ranking queries are essential tools to process these databases and return only the most relevant answers of a query, based on a scoring function. Many approaches were proposed to study and analyze the problem of efficiently answering such ranking queries. Managing distributed uncertain database is also an important issue. In fact ranking queries in such systems are an open challenge. The main objective of this paper is to discuss ranking in distributed uncertain database along with its issued problems. Starting with uncertain data representation, query processing and query types in such systems are discussed along with their challenges and open research area. Top-k query is presented with its properties, as a ranking technique in uncertain data environment, mentioning distributed top-k and distributed ranking problems.

Web-oriented agent-based system for monitoring real-time data processing

Real-time data processing systems generally contain a series of heterogeneous, complex and critical processes. These systems use monitoring tools to increase their productivity and efficiency by detecting failures and tracing workflow progress for the different processes. However, most of the current monitoring tools are platform dependent, process/task specific, expensive, difficult to maintain, and consume many of the organization resources. This paper introduces a prototype generic agent-based and task-unspecific system that overcomes these limitations and restrictions. The proposed platform-independent system is structured into three different modules. These modules collect, organize, infer, and visualize the status of the different objects as well as their necessary attributes/properties from the different stages of the monitored system. The proposed agent-based monitoring system is fast, dynamic, easy to configure, and more convenient to implement. The monitoring agent is reactive, autonomous and communicative. Moreover, it can be used either from any internet browser or in a stand alone mode.

A novel strategy for a vertical web page classifier based on continuous learning naïve bayes algorithm

Abstract Web page classification may be considered as a one of the most challenging research areas. Where the web has a huge volume of unstructured documents of distributed data related to a variety of domains; so, considering one base for the classification task will be extremely difficult. In addition, the web is full of noise that will certainly harm the classifier performance especially if it is found in the classifier training data. Generally, it will be more valued to build a domain-oriented classifiers (vertical classifiers) to classify pages related to a specific domain. This paper analyzes a new way of applying Bayes theorem to build a Domain-Oriented Naive Bayes (DONB) classifier. In addition, a main contribution is to introduce a novel classification strategy by adding the continuous learning ability to bayes theorem to build a Continuous Learning Naive Bayes (CLNB) classifier. Where the overfitting problem has a great impact on most web page classification techniques, continuous learning can be considered as a proposed solution, it allows the classifier to adapt itself continuously for achieving better performance. Both classifiers are tested; experimental results have shown that CLNB demonstrate significant performance improvement over DONB , where its accuracy reaches 94.1% after testing 1000 page. In addition, according to continuous learning, more accuracy enhancement is predicted during future tests.