Zainab Abu Bakar

Constructing Dynamic Infrastructure as a Service Model (DIAAS) According to User Preferences

Cloud computing has forced providers to give satisfying services to users. Unfortunately, there is no model that integrates user preferences with services offered by providers. The services provided are subject to change and thus this information must be reflected in the proposed dynamic model so the user can make an up-to-date decision in choosing the provider based on his preferences. In order to construct and evaluate this dynamic infrastructure as a service (DIAAS) model, the services considered are the speed of central processing unit (CPU), the size of random-access memory (RAM), the size solid-state drive (SSD), the bandwidth in bits per second (bit/s), and the cost of service. The DIAAS uses intelligent tool (ITOOL) for grabbing current provider functional services and stores user preferences. ITOOL retrieves the values of services either by using Web Services or JSON. The services are weighted using linear equations and ranked using average sum of the weighted services. DIAAS will display the list of providers according to user preferences after performing weighting and ranking procedures. There are changes in the value of services by providers, and this implies that user has to be aware of these changes. DIAAS can also be used by providers to improve their services. The findings of DIAAS model will be provided based on three levels (Low, Medium, High). Low = 33.33%, Medium = 6.66%, High = 100%. In DIAAS model, the low percentage will be given to the lower weight of the service and the high percentage of highest weight. Except for cost, the low percentage will be given to highest weight and the high percentage of lower weight. After that, we calculate the weight of each service for each provider by the linear equation formula. Finally, the rank value for each provider is the average of the summation of weights for all the services.

Identifying the Qur’anic Segment from Video Recording

This paper describes a system to identify Quran recitation (referred as Qur’anic) segment from speech video recording using the extracted acoustic signal. Identifying the Qur’anic sequence pattern from mixed-combination of speech and Qur’anic signal will contribute to more efficient segmentation of video segments. The random forest classifier algorithm is employed to classify the dynamic pattern of the extracted audio. Two feature sets which are pitch and intensity are extracted from the audio, and constructed into sequence of speech patterns which then classified as Qur’anic or non-Quranic segments. A collection of 40 segmented videos were trained and compared with the segmented videos which have been segmented manually. This project achieves classification accuracy of 57% using pitch and 85% using intensity. While using pitch feature only, 85% of the identified segments match the manually segmented collection while using intensity feature gives 95% match accordingly).

Prioritization tool of Cloud Computing service provider based on user requirement

The problem arises when there is no tool that can help users to make best decision for choosing provider that meets user requirements satisfactorily. Available tools only compare the services of providers without considering user requirements. Thus the objective of this paper is to present a tool, called PTUR to meet the requirements and satisfaction of user from the selected providers that are ranked according to the users requirements. To fulfill the objectives, the information of user requirements and functional services of providers are gathered. The prominent information required from the providers are the speed of central processing unit (CPU), the size of random-access memory (RAM), the size solid-state drive (SSD), the bandwidth in bits per second (bit/s), the cost of service. Next, the construction of the PTUR that stores of provider functional services and the user requirement needs; performs weighting using linear equation; and ranks the providers by summing the weightage of services for each provider. The results display the ranking of the providers based on user requirements and the user can choose what PTUR recommended or make decision to select other provider. PTUR also saves time for the user and be used by providers to benchmark or improve their services.

Identifying concept from English translated Quran

Ontology learning is a field of extracting ontological elements to form ontology. Identification of concepts is the main activities within ontology learning. Diverse methods can be used to find concepts. One of the methods is using collocation learning technique. The technique used statistical scores which to test the strength of the connection between terms. In English translated Quran, single term Allah has occurred more frequently. The highest occurrences make the term Allah as concept but ignore the multi terms that related terms to Allah. This paper proposed a method to extract concept. It is based on collocation of terms related to Allah. The collocation used Ngram method. The result shows that the collocation method is able to identify terms related to Allah to be as concepts.

Query reformulation using ontology and keyword for durian web search

Query reformulation techniques based on ontological approach have been studied as a method to improve retrieval effectiveness. However, the evaluation of this techniques has primarily focused on comparing the technique with ontology and without ontology. The aim of this paper is to present, evaluate and compare the proposed technique in four different possibilities of reformulation. In this study we propose the combination of ontology terms and keywords from the query to reformulate new queries. The experimental result shows that reformulation using ontology terms alone has increases recall and decreases precision. However, better results were obtained when the ontology terms being combined with the querys keywords.

Prosodic breaks on Malay speech corpus: Evaluation of pitch, intensity and duration

Prosodic phrasing is useful to segment lengthy spontaneous speech into smaller meaningful utterance without analysis of linguistic information. A simpler approach is presented to identify and classify the boundaries in prosodic phrasing using pitch and intensity patterns and pause duration on Malay speech sentences. We also propose a listening test that allows trained listener to classify the boundaries as minor or major breaks. This cheaper and faster approach is proven useful for under-resource language such as Malay which do not have comprehensive prosodic-annotated corpus. Word-related pitch, intensity and duration features are extracted from the targeted sentence and phrase breaks. A speech corpus is developed from targeted breaks of 100 speech sentences evaluated in the listening test. Instead of labeling the phrase break using linguistic and phonetic meaning, the proposed listening test allows labeling of phrase break as perceived by listener. In addition, the results can be used as preliminary information for evaluation of boundary saliency at the targeted boundary locations.

Information extraction: Evaluating named entity recognition from classical Malay documents

Natural Language Processing (NLP) is an important field of research in Computer Science. NLP is the process of analyzing texts based on a set of theories and technologies, and recent studies focused more on Information Extraction (IE). In Information Extraction, there are few steps or commonly known as task to be followed, which are named entity recognition, relation detection and classification, temporal and event processing, and template filling. Recent researches in Malay languages mainly focused on newspaper articles and since this research experiment is experimenting on classical documents, there is a need to identify the best way to extract noun from existing methods. This paper proposes to conduct a research about extracting nouns from Malay classical documents. The result shows that experiment using the Noun Extraction using Morphological Rules (Verb, Adjective and Noun Affixes) that has 77.61% chances of identifying a noun to contribute to the existing Malay noun list. As there is not any existing completed Malay noun list or dictionary that can be used as a guide, the results extracted still need to be judged by the language experts.