Search

The upstream capital project oil and gas industry is considered a critical sector in Malaysia. Apart from its significant monetary contribution to the country, big data analysis is also applied to the supply chain operation. The prescriptive analysis is based on Artificial intelligence (AI), specifically Machine Learning (ML), which involves algorithms and models that enable computers to make decisions based on mathematical data relationships and patterns. This study aims to identify ML analysis in Malaysia’s upstream capital projects, which may improve business decisions via the use of statistical models and ML algorithms. Incorporating ML algorithms and statistical models will produce better business decision-making by enhancing efficiency and productivity besides fast monetisation and minimising risk and returns. Overall, with the use of mixed analysis elements, it can produce better decision support for stakeholders and company owners before making crucial business decisions.

A variety of applications have been built in recent years with the aim to extract knowledge from Al-Quran. Current knowledge representations of Al-Quran give attention primarily on conceptual ontology models that describe the semantic relations between the Quranic concepts or entities. There seems to be minimal effort towards recognizing the semantic relations between words in Quranic text, which is relatively more complex. This paper aims to present a framework for semantic knowledge representation of Al-Quran using dependency relations between words, in an attempt to boost the retrieval accuracy for Al-Quran. The semantic analysis is performed on Quranic verses according to word dependency relations using dependency parsing. Based on parsed dependencies, a set of rules are formulated to build a semantic graph of Surah Ali Imran of Al-Quran. The efficiency of the semantic representation was tested by developing a prototype question answering system. The framework was evaluated using precision and recall, First Hit Success, First Answer Reciprocal Rank and Total Reciprocal Rank by comparing the retrieved and actual answers. The results indicate that the performance of the proposed framework using word dependencies improves the semantic representation of knowledge.

Named Entity Recognition (NER) is an early stage processing of an Information Extraction, which identifies and classify entities in text. The outcomes of the task have become the foundation of building complex Information Extraction applications. With the enormous amount of information available everywhere, the area has gained lots of attention from the research community. Currently, there are two main approaches to perform the Named Entity Recognition; rule-based approach and machine learning approach. In order to improve the accuracy of the classification and performance of the recognizer, some researchers have implemented a hybrid approach, which is the combination of both approaches. Even though many research and works have been done in the Named Entity Recognition, there is still room available for improvement. This paper proposed to increase the accuracy of entities detected by implementing anaphora resolution during the preprocessing phase and a hybrid approach to classify the detected tokens during the classification phase. The hybrid approach is combined the Conditional Random Field (CRF) classifier with a gazetteer and pattern rules to perform classification. The result has shown that the application of anaphora and gazetteer has increased 46% accuracy of the detected entities for the person class.

This paper discussed the idea of the computer system capable of simulating understanding with respect to reading a text document. The research is concerned with the problem of generating sophisticated knowledge representation for the purpose of understanding the natural language. Due to that, a simplification form of logical-oriented model of knowledge representation called Pragmatic Skolem Clauses (PSC) is proposed to represent the semantic formalism for the computational linguistic. Each set of pragmatic skolem clauses containing at least one skolem constant, which shows the thematic role relationship between clauses. Semantically and pragmatically-accented approach will be discussed in this paper in the context of formal grammar and linguistic semantic.

Objectives: This paper describes work in constructing two models of knowledge representation in order to increase the performance of retrieving information on English Quran domain. Due to many approaches available to construct knowledge representation, this paper only focused on ontology-based and graph database-based knowledge representation. Methods: Quranic Arabic Corpus that available at http://www.corpus.quran.com is used as a source to build the knowledge representation. This paper extracted several data from it which include English token, token location, and token Part Of Speech (POS). Protege is used to construct the ontology and Neo4j is utilized in developing the graph database. Findings: This paper has presented a construction process of two models of Knowledge Representation(KR)which are ontology and graph database based KR. These models will be used in developing of an English Quran question answering system in order to discover which model exhibit the best achievement in increasing question answering system performance. Improvements/Applications: The authors hope there will be useful information gained from the question answering system experiments when using both of those models.

This paper describes a work in constructing two models of knowledge representation (KR) in aiming to do evaluation of their achievement in contributing to increase performance of retrieving information on English Quran domain. Due to many approaches available to construct a KR in providing data for information retrieval process, there is a need to find out in what model the KR could provide a valuable contribution for retrieving information. We focused on ontology-based KR and graph database-based KR. We use Quranic Arabic corpus that available at http://www.corpus.quran.com as a source to build the KR. We extracted several data from it i.e. English token, token location, and token Part of Speech (POS). Protégé is used to construct the ontology and Neo4j is utilized in developing the graph database. Both KR models will be equipped in developing of an English Quran Question Answering system in order to evaluate their benefit.

The knowledge behind the gigantic pool of data remains largely unextracted. Techniques such as ontology design, RDF representations, hpernym extraction, etc. have been used to represent the knowledge. However, the area of logic (FOPL) and linguistics (Semantics) has not been explored in depth for this purpose. Search engines suffer in extraction of specific answers to queries because of the absence of structured domain knowledge. The current paper deals with the design of formalism to extract and represent knowledge from the data in a consistent format. The application of logic and linguistics combined greatly eases and increases the precision of knowledge translation from natural language. The results clearly indicate the effectiveness of the knowledge extraction and representation methodology developed providing intelligence to machines for efficient analysis of data. The methodology helps machines to retrieve precise results in an efficient manner.

Morphological analysis is used to study the internal structure words by reducing the number of vocabularies used while retaining the semantic meaning of the knowledge in NLP system. Most of the existing algorithms are focusing on stemmatization instead of lemmatization process. Even with technology advancement, yet none of the available lemmatization algorithms able to produce 100 % accurate result. The base words produced by the current algorithm might be unusable as it alters the overall meaning it tried to represent, which will directly affect the outcome of NLP systems. This paper proposed a new method to handle lemmatization process during the morphological analysis. The method consists three layers of lemmatization process, which incorporate the used of Stanford parser API, WordNet database and adaptive learning technique. The lemmatized words yields from the proposed method are more accurate, thus it will improve the semantic knowledge represented and stored in the knowledge base.

Machine Translation has been defined as the process that utilizes computer software to translate text from one natural language to another. This definition involves accounting for the grammatical structure of each language and using rules, examples and grammars to transfer the grammatical structure of the source language (SL) into the target language (TL). This paper presents English to Arabic approach for translating well-structured English sentences into well-structured Arabic sentences, using a Grammar based and example-translation techniques to handle the problems of ordering and agreement. The proposed methodology is flexible and scalable, the main advantages are: first, a hybrid-based approach combined advantages of rule-based (RBMT) with advantages example-based (EBMT), and second, it can be applied on some other languages with minor modifications. The OAK Parser is used to analyze the input English text to get the part of speech (POS) for each word in the text as a pre-translation process using the C# language, validation rules have been applied in both the database design and the programming code in order to ensure the integrity of data. A major design goal of this system is that it will be used as a stand-alone tool, and can be very well integrated with a general machine translation system for English sentences.

We are in the information intensive environment in which various forms of digital contents have been growing exponentially. In this era of digital data, knowledge representation has been considered as a crucial component of any information retrieval system. It is also considered as a major problem especially in representing the content of unstructured text in an effective way. Although the mission remains impossible to achieve 100% accuracy, many researchers are indulging themselves in documenting these data in many different techniques so that it can be communicated effectively and easily. Indexing is an important element that determines the success of retrieval. Since we are dealing with multiple documents, preprocessing of data is needed before the data gets indexed. Thus, this paper presents an approach on the preprocessing technique. The semantic data which have been represented in skolem clauses will be preprocessed with the help of automatic lexicon generator output and WordNet. This preprocessing plays an important role in getting rid of redundant data before it gets indexed into the semantic matrix. Besides redundancy, it also helps in dealing with common problem that exists in indexing multiple documents in which similar sentences with more or less the same meaning but have been constructed by using different sets of words. As a conclusion, the integration of WordNet and lexicon leads to better result in terms of building effective knowledge representation.

There have been very few studies on the use of conflation algorithms for indexing and retrieval of Malay documents. The two main classes of conflation algorithms are string-similarity algorithms and stemming algorithms. Stemming is used in information retrieval systems to reduce variant word forms to common roots in order to improve retrieval effectiveness. As in other languages, there is a need for an effective stemming algorithm for the indexing and retrieval of Malay documents. Again there are few research on n-gram string-similarity measures done on Malay. We have experimented on the application of stemming and string similarity matching on retrieving of verses from the Al-Quran in order to evaluated their effectiveness. Before retrieval effectiveness can be carried out an experimental data set need to be developed which comprises of a collection of documents, a set of queries, and their relevant judgements. In this paper we will describe the development of the experimental data set and the application of stemming and similarity matching algorithms in retrieving the verses from the Al-Quran. Inherent characteristics of n-grams and several variations of experiments performed on the queries and documents are discussed. The variations are: both non-stemmed queries and documents; stemmed queries and nonstemmed documents; and both stemmed queries and documents. Further experiment are then carried out by removing the most frequently occurring n-gram. The dice-coefficient is used as threshold and weight in ranking the retrieved documents. Beside using dice coefficients to rank documents, inverse document frequency weights are also used. Interpolation technique and standard recall-precision functions are used to evaluate the retrieval effectiveness.

The representation of document content is very important factors in retrieval process. The failure to create a good knowledge representation will definitely lead to failure in terms of its retrieval no matter how good the retrieval engine is. Therefore, this research focused on creating a reliable knowledge representation for our retrieval engine. We are using skolem to capture the information conveyed by multiple text documents and used skolem as an index language. This research also focuses on utilizing the skolem index as its knowledge representation in its question answering system. The system is capable of retrieving the answer as well as states the exact document in which the answer is derived from.

Region growing and edge detection are two popular and common techniques used for image segmentation. Region growing is preferred over edge detection methods because it is more robust against low contrast problems and effectively addresses the connectivity issues faced by edge detectors. Edgebased techniques, on the other hand, can significantly reduce useless information while preserving the important structural properties in an image. Recent studies have shown that combining region growing and edge methods for segmentation will produce much better results. This paper proposed using edge information to automatically select seed pixels and guide the process of region growing in segmenting geometric objects from an image. The geometric objects are songket motifs from songket patterns. Songket motifs are the main elements that decorate songket pattern. The beauty of songket lies in the elaborate design of the patterns and combination of motifs that are intricately woven on the cloth. After experimenting on thirty songket pattern images, the proposed method achieved promising extraction of the songket motifs.

The rapid growth of digital data and users’ information needs have made the demands for automatic indexing to become more important than before. Indexing based on keyword has proven to be unsuccessful to cater for the current needs. Thus, this paper presents a new approach in creating semantic skolem indexing for multiple documents that automatically index all the documents into single knowledge representation. The skolem indexing matrix will then be incorporated in question answering system to retrieve the answer for users query.

Thesaurus has become another valuable structure in any Information Retrieval system. It is a list of terms and concepts that provide a controlled vocabulary of words to use in document indexing, clustering, searching and retrieval. This paper present the results of expanding user's query using Malay thesaurus in the process of searching Malay documents from Malay Hadith retrieval system. The results obtained shows that the retrieval effectiveness improves by four percent when thesaurus is employed in the process of retrieving Malay translated Hadith documents, compared to when single term queries are employed.

Over the past decades, computer revolution has opened up many possibilities for new field of investigation. With greater accessibility to information and lowering cost of powerful computers, this has spawned new efforts towards understanding complex tasks. Lexicon in particular has long been recognized as interesting and challenging because of its complexness. It is the knowledge of individual words in the language that has been perceived as central component for all types of natural language processing system. In this paper we present an algorithm to create an automatic lexicon generator in order to generate lexicon from an input document by making use of Apple Pie Parser. The lexicon generated managed to reduce significant amount of time and manpower drastically. Psycholinguists as well as computational linguists can benefit from this automatic lexicon construction.

Even though researches in content-based image retrieval are one of the most intensive research areas of information retrieval, they almost certainly focused on retrieving natural images or photographs. Artificially-produced images are stylized form of natural, real-life objects. Research for these images have received less attention, thus this paper attempts to intensify work in this area by implementing quantitative and qualitative performance evaluation of artificially-produced images. Fifty selected images are used as sample queries and they are retrieved based on fusion of geometric shape descriptors. Similarities of these images are measured using Euclidean distance and performances of the retrievals are evaluated using both quantitative and qualitative tests. Quantitative test using recall-precision rate and hit-miss ratio showed that geometric shape descriptors are effective in retrieving the images. However, qualitative test results demonstrate agreement between the system and the interviewed people in the assignment of similarity ranks is at average level.

Computer science is in the more challenging era due to the digital growth and demand that we are facing today. A typical IR system will not go far enough as it uses keyword in order to retrieve the desired information. On the other hand, natural language question answering which is based on logic retrieval have proven to perform better than other methods. The most important component of question answering system is the lexicon. Researchers have identified this as the backbone of any NLP system. In this paper we present our approach on how an automatic lexicon generator has been integrated with question answering system which uses logical inference model. This paper also shows on how the automatic lexicon generator has helped in creating knowledge representation to be used by logic based question answering system. Besides that, the lexicon generated managed to reduce significant amount of time and manpower. It has also helped to make the question answering system to be more robust.

Support vector machines (SVMs) and related kernel methods have become widely known tools for text mining tasks such as classification and regression. The Arabic part of speech (POS) based support vectors machine is designed and implemented. The NeuroSolutions software is used to adopt and learn the proposed tagger. The radial basis functions (RBFs) is used as a linear function approximator. The experiments has give an evinced that the SVMs tagger is accurate of (99.99%), has low processing time, and use a little a mount of data at training phase.

Machine Translation has been defined as the process that utilizes computer software to translate text from one natural language to another[2].This definition involves accounting for the grammatical structure of each language and using rules and grammars to transfer the grammatical structure of the source language (SL) into the target language (TL). Agreement and Word-ordering play an important role with this structure, as MT should handle agreement between the subject and verb where the number gender, person and features of the subject are important factors in the derivation of the verb as well as the features of the verb itself. Other agreements are required between the adjective and the noun where Arabic adjectives depend on the number, gender and person as well as the definiteness and indefiniteness of the noun. Some other agreements also exist between the numbers and the countable nouns. This paper examines the agreement and word-order rules in tackling Arabic MT systems. Special attention is given to Tarjim by Sakhr Software Company which pioneered a new generation of Arabic Natural Language Processing (NLP) technologies. The approach adopted is flexible and scalable, and the main advantages are: first, it is a rule-based approach, and second, it can be applied on other languages with minor modifications. A major design goal of this system is that it can be used as a stand-alone tool and can be very well integrated with a general machine translation system for English sentence.

Mathematical morphological operations are commonly used as a tool in image processing for extracting image components that are useful in the representation and description of region shape. In this paper, six basic morphological operations are investigated to remove noise and enhance the appearance of binary images. Dilation, erosion, opening, closing, fill and majority operations are tested on twenty-five images and subjectively evaluated based on perceived quality of the enhanced images. Results of the experiments showed that noise can be effectively removed from binary images using combinations of erode-dilate operations. Also, the binary images are significantly enhanced using combinations of majority-close operations.

The idea of the computer system capable of simulating understanding with respect to reading a story or passage and answering questions pertaining to it, has received increased attention within the NLP community as a means to develop and evaluate robust question answering methods. This research is concerned with the problem of generating an automated answer in the context of sophisticated knowledge representation, reasoning, and logical inferential processing. The research focused on Wh types of question to a restricted domain. External knowledge sources i.e. world knowledge (WK) and hypernyms matching procedure (HMP) were introduced. World knowledge led to the refinement of the ability of the system to extract the relevant answers. It provided a solution to the outstanding problem related to the ambiguity enclosed by anaphora and synonym words. Meanwhile, hypernyms refer to a more general or broad keywords, which was able to widen the results of the search. Thus, hypernyms matching procedure was to give a more coherent meaning of words, and eased in the process of extracting the answer based on a given question. This research found that the combined external knowledge sources on word dependencies improved the accuracy of the question answering system with respect to human performance measures.

This paper introduces a logical-linguistic model of document retrieval systems and describes an implementation of a system called SILOL which is based on this model. SILOL uses a shallow semantic translation of natural language texts into a first order predicate representation in performing a document indexing and retrieval process. Some preliminary experiments have been carried out to test the retrieval effectiveness of this system. The results obtained show improvements in the level of retrieval effectiveness, which demonstrate that the approach of using a semantic theory of natural language and logic in document retrieval systems is a valid one.

Songket is a traditional handwoven cloth of the Malays and its beauty lies on the design of the songket motifs intricately woven on the cloth. Even though color is a feature of the motif, shape is more important in determining the type of the motifs. This paper describes and tests the efficiency of retrieving songket motifs images using different combination of five geometric shape descriptors. Even though there are many existing approaches of shape-based image retrievals, the goal of this paper is to take advantage of the faster calculations of the geometric shape descriptors to retrieve the images. Fifty selected songket motifs are used as sample queries in this paper. Similarities of these images are measured using Euclidean distance and performance of the retrieval effectiveness is evaluated using recall and precision rate. Results of the experiment showed that geometric shape descriptors are effective in retrieving the songket motifs.

This paper discussed the effectiveness of several popular gradient-based edge detection techniques when applied on binary images of songket motifs. Five edge detector algorithms that is Roberts, Sobel, Prewitt, Laplacian of Gaussian and Canny are applied to twenty-five Malaysian traditional songket motifs. These scanned motif images are initially preprocessed to remove noise using several morphological operations. Other than noise removal, binarization of color images are also done to produce binary images. To determine the performance of the edge detectors, the results are evaluated by five human subjects based on several pre-conceived criteria.

The levels-of-processing theory proposes that there are many ways to process and code information. The level of processing adopted will determine the quality of the representation used to store the information in the computer memory or storage. The levels-of-processing applied in information retrieval can be classified as follows: string processing, morphological processing, syntactic processing and semantic processing. These level-of-processing are imbedded into various models of information retrieval. Conventional information retrieval models, such as Boolean and vector space models rely on an extensive use of keywords, as independent strings, and their frequencies in storing and retrieving information. Thus string processing and morphological processing are mainly adopted in these models. It is believed that such an approach has reached its upper limit of retrieval effectiveness, and therefore, new approaches should be investigated for the development of future systems that will be more effective. With current advances in programming languages and techniques, natural language processing and understanding, and generally in the fields of artificial intelligence and cognitive science, there are now attempts made to include knowledge representation and linguistic processing into information retrieval systems. We also focus our research on the application of certain techniques on specific languages. Besides English, we focus the application of certain techniques especially on Malay. In this paper we will highlight some of the research done in the area of information retrieval at the various levels of processing, and also expound the current research we are doing and the future direction that we would like to undertake.

This paper concerns an application of latent semantic indexing on Malay and English cross language information retrieval system. The retrieval effectiveness was tested on the actual Quranic collection using latent semantic indexing model. The results show that average precision on cross language is higher than monolingual retrieval.