Nasir Sulaiman

Modification of standard function point complexity weights system

Function Point (FP) is a software size measure, which includes the standard FP and many different models derived from it. The standard FP method created by Albrecht in 1979 is currently known as the International FP User group (IFPUG) version, which consists of three main parts: The first part is five components, and the second is the complexity weights that include three levels of complexity; simple, average, and complex. The third part is the general system characteristics of software projects, which consists of 14 technical complexity factors. Although, FP was widely used as a software size measure, but it still suffers from many weaknesses. One of which is the subjectivity in the weights system. In this paper a new FP weights system was established using Artificial Neural Networks. This method is a modification of the complexity weights of FP measure (IFPUG version). The final results were very accurate and much suitable when they were applied on real data sets of software projects.

A review: accuracy optimization in clustering ensembles using genetic algorithms

The clustering ensemble has emerged as a prominent method for improving robustness, stability, and accuracy of unsupervised classification solutions. It combines multiple partitions generated by different clustering algorithms into a single clustering solution. Genetic algorithms are known as methods with high ability to solve optimization problems including clustering. To date, significant progress has been contributed to find consensus clustering that will yield better results than existing clustering. This paper presents a survey of genetic algorithms designed for clustering ensembles. It begins with the introduction of clustering ensembles and clustering ensemble algorithms. Subsequently, this paper describes a number of suggested genetic-guided clustering ensemble algorithms, in particular the genotypes, fitness functions, and genetic operations. Next, clustering accuracies among the genetic-guided clustering ensemble algorithms is compared. This paper concludes that using genetic algorithms in clustering ensemble improves the clustering accuracy and addresses open questions subject to future research.

OPWUMP: An Architecture for Online Predicting in WUM-Based Personalization System

The Internet is one of the fastest growing areas of intelligence gathering. During their navigation web users leave many records of their activity. This huge amount of data can be a useful source of knowledge. Sophisticated mining processes are needed for this knowledge to be extracted, understood and used. Web Usage Mining (WUM) systems are specifically designed to carry out this task by analyzing the data representing usage data about a particular Web Site. WUM can model user behavior and, therefore, to forecast their future movements. Online prediction is one web usage mining application. However, the accuracy of the prediction and classification in the current architecture of predicting users’ future requests systems can not still satisfy users especially in Huge Web sites. To provide online prediction efficiently, we develop an architecture for online predicting in WUM-based personalization system (OPWUMP).This article advances an architecture of Web usage mining for enhancing accuracy of classification by interaction between classification, evaluation, current user activates and user profile in online phase of this architecture.

Data stream clustering by divide and conquer approach based on vector model

Recently, many researchers have focused on data stream processing as an efficient method for extracting knowledge from big data. Data stream clustering is an unsupervised approach that is employed for huge data. The continuous effort on data stream clustering method has one common goal which is to achieve an accurate clustering algorithm. However, there are some issues that are overlooked by the previous works in proposing data stream clustering solutions; (1) clustering dataset including big segments of repetitive data, (2) monitoring clustering structure for ordinal data streams and (3) determining important parameters such as k number of exact clusters in stream of data. In this paper, DCSTREAM method is proposed with regard to the mentioned issues to cluster big datasets using the vector model and k-Means divide and conquer approach. Experimental results show that DCSTREAM can achieve superior quality and performance as compare to STREAM and ConStream methods for abrupt and gradual real world datasets. Results show that the usage of batch processing in DCSTREAM and ConStream is time consuming compared to STREAM but it avoids further analysis for detecting outliers and novel micro-clusters.

The learning of an opponent's approximate preferences in bilateral automated negotiation

Autonomous agents can negotiate on behalf of buyers and sellers to make a contract in the e-marketplace. In bilateral negotiation, they need to find a joint agreement by satisfying each other. That is, an agent should learn its opponents preferences. However, the agent has limited time to find an agreement while trying to protect its payoffs by keeping its preferences private. In doing so, generating offers with incomplete information about the opponents preferences is a complex process and, therefore, learning these preferences in a short time can assist the agent to generate proper offers. In this paper, we have developed an incremental on-line learning approach by using a hybrid soft-computing technique to learn the opponents preferences. In our learning approach, first, the size of possible preferences is reduced by encoding the uncertain preferences into a series of fuzzy membership functions. Then, a simplified genetic algorithm is used to search the best fuzzy preferences that articulate the opponents intention. Experimental results showed that our learning approach can estimate the opponents preferences effectively. Moreover, results indicate that agents which use the proposed learning approach not only have more chances to reach agreements but also will be able to find agreements with greater joint utility.

Artificial Immune Recognition System with Nonlinear Resource Allocation Method and Application to Traditional Malay Music Genre Classification

Artificial Immune Recognition System (AIRS) has shown an effective performance on several machine learning problems. In this study, the resource allocation method of AIRS was changed with a nonlinear method. This new algorithm, AIRS with nonlinear resource allocation method, was used as a classifier in Traditional Malay Music (TMM) genre classification. Music genre classification has a great important role in music information retrieval systems nowadays. The proposed system consists of three stages: feature extraction, feature selection and finally using proposed algorithm as a classifier. Based on results of conducted experiments, the obtained classification accuracy of proposed system is 88.6 % using 10 fold cross validation for TMM genre classification. The results also show that AIRS with nonlinear allocation method obtains maximum classification accuracy for TMM genre classification.

MALIM — A new computational approach of malay morphology

Malay is categorized as an Austronesian language, a group which also contains Bahasa Indonesia and Tagalog. Quite number of morphological analyzers has been developed for Malay, including based on two-level formalism, stemming/conflation model, or even specific model. The obvious weaknesses are incompleteness and incapability of handling ambiguity which affect the accuracy of analysis. So we introduced a new technique called S-A-P-I to handle them in our analyzer — MALIM. In this paper we describe about MALIM and the empirical study to its usage. Our results proved that by using our technique increase the accuracy of morphological analysis up to 98.99% which covers 99.99% of sample data. Thus we believe this approach is the most suitable in handling morphological analysis for Malay.

Quorum-based Data Replication in Grid Environment

Replication is a useful technique for distributed database systems and can be implemented in a grid computation environment to provide a high availability, fault tolerant, and enhance the performance of the system. This paper discusses a new protocol named Diagonal Data Replication in 2D Mesh structure (DR2M) protocol where the performance addressed are data availability which is compared with the previous replication protocols, Read-One Write-All (ROWA), Voting (VT), Tree Quorum (TQ), Grid Configuration (GC), and Neighbor Replication on Grid (NRG). DR2M protocol is organized in a logical two dimensional mesh structure and by using quorums and voting techniques to improve the performance and availability of the replication protocol where it reduce the number of copies of data replication for read or write operations. The data file is copied at the selected node of the diagonal site in a quorum. The selection of a replica depends on the diagonal location of the structured two dimensional mesh quorum where ...

Proactive architecture for Internet of Things (IoTs) management in smart homes

Smart homes are driven by heterogeneity in nature and consist of diverse components that promote user comfort and security. In recent times, tremendous growth of Internet of Things (IoTs) applications is seen in smart homes. The huge diversity of various IoTs applications generally leads to interoperability requirements that need to be fulfilled. Current IoTs management is achieved using physical platforms that lack intelligence on decision making. A proactive architecture that deploys Event-Condition-Action (ECA) method is proposed to resolve the management of heterogeneous IoTs in smart homes. The proactive architecture, developed with a core repository stores persistent data of IoTs schema, proved to be an ideal solution in solving interoperability in smart homes.

Activity recognition based on accelerometer sensor using combinational classifiers

In recent years, people nowadays easily to contact each other by using smartphone. Most of the smartphone now embedded with inertial sensors such accelerometer, gyroscope, magnetic sensors, GPS and vision sensors. Furthermore, various researchers now dealing with this kind of sensors to recognize human activities incorporate with machine learning algorithm not only in the field of medical diagnosis, forecasting, security and for better live being as well. Activity recognition using various smartphone sensors can be considered as a one of the crucial tasks that needs to be studied. In this paper, we proposed various combination classifiers models consists of J48, Multi-layer Perceptron and Logistic Regression to capture the smoothest activity with higher frequency of the result using vote algorithmn. The aim of this study is to evaluate the performance of recognition the six activities using ensemble approach. Publicly accelerometer dataset obtained from Wireless Sensor Data Mining (WISDM) lab has been used in this study. The result of classification was validated using 10-fold cross validation algorithm in order to make sure all the experiments perform well.