Indigenous Food Recognition Model Based on Various Convolutional Neural Network Architectures for Gastronomic Tourism Business Analytics

In gastronomic tourism, food is viewed as the central tourist attraction. Specifically, indigenous food is known to represent the expression of local culture and identity. To promote gastronomic tourism, it is critical to have a model for the food business analytics system. This research undertakes an empirical evaluation of recent transfer learning models for deep learning feature extraction for a food recognition model. The VIREO-Food172 Dataset and a newly established Sabah Food Dataset are used to evaluate the food recognition model. Afterwards, the model is implemented into a web application system as an attempt to automate food recognition. In this model, a fully connected layer with 11 and 10 Softmax neurons is used as the classifier for food categories in both datasets. Six pre-trained Convolutional Neural Network (CNN) models are evaluated as the feature extractors to extract essential features from food images. From the evaluation, the research found that the EfficientNet feature extractor-based and CNN classifier achieved the highest classification accuracy of 94.01% on the Sabah Food Dataset and 86.57% on VIREO-Food172 Dataset. EFFNet as a feature representation outperformed Xception in terms of overall performance. However, Xception can be considered despite some accuracy performance drawback if computational speed and memory space usage are more important than performance.

Free Open-Source High – Availability Solution for Java Web Application Using Tomcat And MySQL

With the growth of the internet, the number of web applications is also growing. Many web applications are becoming more important to the stakeholders that they cannot afford downtime which can cause loss of revenue, loss of productivity, etc. In the past, only big organizations with deep pocket could afford implement high-availability for their web application, but nowadays there are free open-source software programs that support high-availability feature available to everyone. This research studied the feasibility of implementing high-availability for Java web application system without using commercial software. This research compared the capability of proprietary and free open-source high-availability solution for Java web application based on a simple high-availability design, where a test Java web application was deployed into the environment based on proprietary and free open-source solutions, and tested how well each solution perform when problem occurs. The result showed that the free open-source high-availability solution worked, but not as well as proprietary one. However, the proprietary high-availability solution for database did not perform well, and neither did the open-source one. This research concludes that the free open-source high-availability solution works and thus made high-availability become much more affordable, especially for individual or small organizations with budget constraints.

A feedback survey on the clinical performance and acceptability of levonorgestrel releasing intrauterine system (EMILY®) during two years of usage

<p class="abstract"><strong>Background:</strong> Objective of the study was to evaluate the long term clinical performance and acceptability of levenorgestrel releasing intrauterine system (LNG-IUS), ‘Emily’, for the treatment of abnormal uterine bleeding (AUB).</p><p class="abstract"><strong>Methods:</strong> A two year feed back survey was carried out by telephonic conversation and the data obtained was fed in to the house web application system in the respective Emily users. Subjective perceptions were verbally assured based on a questionnaire framed on the quality of the life led by the Emily users. </p><p class="abstract"><strong>Results:</strong> The treatment resulted in a significant reduction in the menstrual bleeding of 102 patients who completed the two year study. The mean bleeding score and SD 18.79 (13.36) in first year falls to 10.76 (12.40) in second year. 78 (76.47%) users were very satisfied with the product usage.</p><p class="abstract"><strong>Conclusions:</strong> Use of ‘Emily’ LNG-IUS resulted in significant reduction of menstrual bleeding in women with AUB in two years and most of the users were very satisfied with the product.</p>

A fill-in-blank problem workbook for Java programming learning assistant system

Purpose To advance Java programming educations, the authors have developed a Java Programming Learning Assistant System (JPLAS) as a web application system. JPLAS provides fill-in-blank problems for novices to study the grammar and basic programming skills through code reading. Design/methodology/approach To select the blank elements with grammatically correct and unique answers from a given Java code, the authors have proposed the graph-based blank element selection algorithm. Then, the code for this problem should be highly readable to encourage novice students to study it. Because code readability can be improved by proper names for identifiers, the authors have also proposed the naming rule testing algorithm to identify codes with correct names. Findings In this paper, the authors present a fill-in-blank problem workbook by collecting suitable Java codes from textbooks and Web sites and applying these algorithms with several extensions. The workbook consists of 16 categories with a considerable number of problems that follow the conventional learning order of Java programming. Originality/value The proper set of ready-made fill-in-blank problems is effective in enhancing the usability of JPLAS both for teachers and students. For the preliminary evaluation, the authors assign a few problems to students. In coming semesters, the authors will use this workbook in the course to verify the adequacy of the proposal for novices.