This paper focuses machine learning techniques Naïve Bayesian Classifier method for the imputation of missing values in large datasets. Single value imputation produce biased results, whereas multiple imputation generates right value to replace. The study was conducted to implement several algorithms in unsupervised techniques like Mean, Median, Standard Deviation, and Regression and supervised machine learning techniques like Naïve Bayesian classifier provides different methods to calculate Class prior probability, Posterior probability, Predictor probability and Maximum Likelihood. Both methods are trying to find hidden structure in unlabeled data. The performance of this method has been compared by using Correlation statistics analysis which produces the imputed values are positively related or negatively related or not related with each other. To evaluate the performance, the standard machine learning repository dataset has been used. We experimentally shows that our approach significantly outperforms some standard machine learning methods for handling missing values in classification tasks.
Brunei Darussalam government is putting in efforts to ensure every Muslims regardless citizens or tourists to be well informed on “Halalan Toyyiban” food and beverages. As ‘Halal’ is a priority for all Muslims, the abused of ‘Halal’ word equally misused by Muslims and Non-Muslims is not to be addressed lightly. Therefore, in order to provide solutions for the issues, a study is conducted with primary objectives to identify ‘Halal’ certified restaurants in Brunei Darussalam and to develop a prototype of ‘Halal’ Verification System using Augmented Reality (AR) Technology. A list of ‘Halal’ certified restaurants is obtained from Brunei Halal Food Council that is utilized as a guideline for conducting observation and questionnaire survey on every ‘Halal’ certified restaurant along with its coordinates in Brunei Darussalam. Augmented Reality feature within Android environment is the basis for the proof of concept. An informative, user friendly and interactivity are the main features of the application that functions mainly to identify ‘Halal’ certified restaurants within user current location. Tests are made with 10 users of smart phone in locating nearest ‘Halal’ certified restaurants. 107 ‘Halal’ certified restaurants in Brunei Darussalam are used as the data set for the prototype’s application development. All tests using the prototype’ application is proven success with users able to locate ‘Halal’ certified restaurants and information pertaining to it. This is aims to promote Brunei Darussalam image as a Muslim’s country as well as Islam as a way of life is further promote.
In any organization, the appropriate use of information systems will have an impact on the effectiveness of day-to-day operations. To find out how the performance of an information system is necessary to do an evaluation of the implementation of which has been run. This study evaluates the implementation of information systems in the Maternity Hospital St. Yosef. By adopting PIECES framework as an evaluation, primary data collection is done through questionnaires distributed. Of the 77 respondents who collected the data used factor analysis to get the factors and indicators that affect the successful implementation of the system and get the right model for the evaluation of the Maternity Hospital in St. Joseph. The results showed that the factor of ease of use and suitability to the needs of a positive effect on the success, while the reliability of the system negatively affect the success. On a scale of 0-100 obtained a minimum of system performance is at 45.75 and the maximum performance of the system is equal to 83.77.
This paper examined the quality of teachers distributed to Secondary Schools in the South-South Geo-political zone of Nigeria. The population of the study comprised of 222,238 teachers adopted from the statistics of teachers in all Secondary Schools in the six Geo-political zone in Nigeria. A purposive sampling technique was adopted to extract a sample size of 44,934 teachers listed by their qualifications and gender in the South-South Geo-political zone. Two research questions were raised to give direction to the study. An expost-facto research design was adopted. Simple percentage was adopted and data presented in tabular form. Findings showed that 46.9% of distributed teachers had degrees with teaching qualification while 22.9% had degrees but do not have teacher qualification. 62% of the total number of teachers distributed to Secondary Schools in the zone were males while 38% were female teachers. Other notable findings were River State’s with the highest number of qualified teachers (60.3%) while Edo State had the least number of qualified teachers (38.6%). Arising from the findings, some recommendations were made; that all untrained degree holders should be encouraged to obtain teaching qualification; that more female teachers should be employed to bridge the gap in the disparity between the number of male to female teachers distributed to Secondary Schools in the area of study and that more teachers should be employed and deployed based on needs and not on selective posting reserved for privileged few.
In Bioinformatics field, DNA sequence alignment had been a long recognized as the main system for comparison methods. The DNA sequence alignment is known to find the optimal alignment. The optimal alignment is the path that search through between the two or more DNA sequences (query sequence (Nq) and subject sequence (Ns)). Recently, the DNA sequences database has showing the limitation in comparisons methods. As the result, the DNA sequence alignment system demands in many platform such as an advanced with high computational performance, high sensitivity and accuracy in methods, low cost and portable system. This paper proposed a software version of Smith-Waterman (SW) algorithm using Field Programmable Gate Array (FPGA). Thirty eight tests with ranging from 1 to 2048 base-pair were conducted to measure software version runtime based. The implementation was carried out using low cost EP4CE115F29C7 FPGA. Thirty eight tests were conducted with the average runtime per cell for 100MHz and 50MHz are recorded at 0.021447ms and 0.044553ms per cell. Therefore, the software implementation has direct dependencies over cell runtime due the iterative computational method used