Sofianita Mutalib

Towards emotional control recognition through handwriting using fuzzy inference

Emotion control is one of personality characteristics that can be detected through handwriting or graphology. One of the advantages is it may help the counselor that has difficulties in identifying the emotion of their counselee. This study is to explore the fuzzy technique for feature extraction in handwriting and then identify the emotion of person. This study uses baseline or slope of the handwriting in determining the level of emotion control whether it is very low, low, medium, high or very high, through Mamdani inference.

Exploring Feature Selection and Support Vector Machine in Text Categorization

With the growing number of text documents in the Internet, it is difficult for users to search, find, manage and organize information quickly. Normally, text documents are classified manually and it is time-consuming. Text categorization is a process of assigning text documents into a set of fixed predefined categories. The high dimensionality of text documents made it difficult to categorize because text documents contain noise and useless data. This paper explored several methods of feature selection that can be used to reduce high dimensionality of feature space in text documents such as Information Gain, Gain Ratio, CHI-Squares, Mutual Information and Document frequency. Next, the study adopted text categorization using Support Vector Machines. The results showed that Support Vector Machines perform well and very fast both in training and testing datasets.

Intelligent Water Dispersal Controller: Comparison between Mamdani and Sugeno Approaches

This paper presents a comparison of fuzzy inference methods in intelligent water dispersal controller primarily focuses on grass watering. In irrigation system, measuring and monitoring soil moisture from the soil information and climatologic factors would determine the amount of water for sufficient soil moisture. Mamdani-style and Sugeno-style inference methods have been tested and evaluated using this information. These methods were tested on normal subsets. Fuzzy rules were determined based on three inputs namely; Bermuda Turf grass coefficient, evapotranspiration (FT) rate, and tensiometer data. The result illustrated that the most convincing fuzzy inference method applied was the Mamdani-style compared to Sugeno-style. It was shown that the controller used less water in turf grass irrigation. Overall, both of the tested methods give significant result to the recognition of soil moisture level.

Dermatology diagnosis with feature selection methods and artificial neural network

Dermatology or skin disease is one of the popular diseases among other diseases these days. The features similarities between different types of skin diseases make diagnosis of skin diseases very complex. A patient needs dermatologist that has a sound and vast good experience in skin diseases in order to give precise results at the right time. This paper elaborates a prototype with back propagation neural network (BPNN) to assist the dermatologist. This prototype improves expert diagnosis method in term of time efficiency and diagnosis accuracy. The use of two feature selection methods namely Correlation Feature Selection (CFS) and Fast Correlation-based Filter (FCBF) help by providing a smaller number of features with greater accuracy and faster response time. The adjustment of parameter in BPNN gives good performance. The findings show that FCBF method offers the shortest elapsed time and highest result compared to CFS method and the full features with an accuracy of 91.2%.

Online Slant Identification Algorithm Using Vector Rules

Signatures are among the most widely accepted personal attributes for identity verification. There are a lot of features that can be discovered in signature which are either dynamic or static features type. An algorithm needs to be designed to extract these signature features. Online system uses pressure sensitive tablets to capture signature of individual as they sign thus analysis can be done directly and immediately. This research explored slant feature algorithm since signature is usually slanted due to the mechanism of handwriting and the human personality. The proposed algorithm are used to formulate the Signature Extraction Features System (SEFS) which provides a set of tools that allow the users to extract slant features in signature automatically for analysis purposes. Twenty individuals from different background are randomly selected to have their signature taken. Their signatures are captured on a tablet and the SEFS would than gather and store the raw data. The image of the signature that is created by the SEFS would be used as samples for the questionnaire to identify the features of slant, where the questionnaires are given to human expert for evaluation. The results from the SEFS are compared with the result from the questionnaire. Results produced by the algorithm for slant extraction shows 85% identical answers compared to the outcome by human expert. These show that the algorithm proposed are promising for further exploration.

Plant Selection System

Expert in agriculture field is in demand, so storing their knowledge would be useful. This research is to analyze the information in agriculture by constructing rules in evaluating the potential plants based on soil suitability and also marketability, yields, consumption or budget. The expertpsilas knowledge was extracted from an agriculture officer and represented in the IF-THEN rules using confident factor. The suggestion is created based on the percentage that was above 80%. In conclusion, the system can give explanations about potential plants, the financial and marketing support and fertilization. It is recommended to apply hybrid expert system and others in future, in order to get even better and accurate results.

Mining Frequent Patterns for Genetic Variants Associated to Diabetes

Data mining consists of crucial tasks in discovering knowledge and hidden patterns and the tasks are significant in the various areas, such as marketing, biomedical, drugs design, event sequences and etc. Frequent pattern mining is a method that has been explored by a lot of researches in discovering new or hidden knowledge. Therefore, this research attempts to see whether frequent pattern mining method could produce significant information from genetic variants by mining deoxyribonucleic acid (DNA) in particular Single Nucleotide Polymorphism (SNP). The experiments were done using sample enumeration algorithm on diabetes data. Based on our experiments, the genetic variants with diabetes risks were found in low support value. The patterns generated were informative to draw relations between the reported risky SNPs with other unreported SNPs.

Towards applying associative classifier for genetic variants

With the availability of biological data and the power of sharing, it produces many opportunities for computer scientists to perform researches in bioinformatics. Generally the researches propose methods for different tasks, mainly to develop algorithms in diagnosing and identification of diseases. One of the primary studies that relevant to health and diseases is genome wide association studies (GWAS). Normally the studies are conducted in different populations to replicate the risk loci of specific disease and the number of groups are keep on progressing, including those from Asian country. Computer scientists should be involved in GWAS due to certain problems and the complexity of the processes involved. The problems and past studies related to GWAS are presented in this paper.

A brief survey on GWAS and ML algorithms

Nowadays, we can see an increasing number of studies in genomics that try to find out ways to detect diseases and also better prevention methods. The public would gain a lot of benefits from the studies. With the rapid development of genotyping technology, it creates opportunity to the researchers to go depth to the genetic and look into the variants. Most of the time, researchers would found different set of variants that increase the risk to the different diseases. Moreover, it is found that different populations would have same or would have different set of variants. The association of the variants to the disease is still in mystery but could be discovered by thorough studies. The studies about the variants are also known as genome wide association studies (GWAS). Key roles in GWAS are not limited to the bioinformaticians or pure scientists only, but also computer scientists could contribute to the studies by developing algorithms and tools. Therefore, this paper would like to briefly introduce GWAS and facilitate researchers with several studies that have applied machine learning (ML) algorithms in GWAS.

Soil classification: An application of self organising map and k-means

This paper discusses the application of two unsupervised methods in classifying type of soils. Soils that are suitable for agricultural activities can be classified into four classes which are hill soil, organic soil, alteration soil and alluvium soil. In addition, no specific support system is able to classify the type of soil and retrieve the information for location and suitable plants for local purposes. In this study, we applied self organizing map (SOM) and k-means in constructing the classification model. The inputs for this study are color, texture, drainage class and terrain. Throughout the process of training and testing, the classification rate for this SOM and k-means are 91.8% and 79.8% respectively.