Search published articles


Showing 5 results for Classification

Ali Yaran,
Volume 18, Issue 6 (7-2007)
Abstract

This paper will promote many questions about the design validity of these exciting blocks. The physical design of these blocks is prone to have social and economical problems, and it is part of the visual pollution and disturbance created on the sky line of Scottish cities special Glasgow. This paper focuses on the necessity to establish a source of initial design for high rise housing design problem classes theory. A thorough research of all the estates in Glasgow area was launched. Thus, a high amount of information was classified, in an attempt to establish a theoretical approach to support the evaluation and appraisal of current blocks problems. The new class's theory in high rise housing design problems is used as a vehicle for conducting this paper analysis. Classification includes physical and social design problems domains. The physical problems domain variables of the numbers of stories, dwellings, entrances, type of corridor, access from the streets and sharing of the site. Furthermore, to the social problems domain variables of graffiti, damage, litter, crime, drugs, child density and health have been examined. Moreover, their interdependent relationships were explored throughout the new classes' theory. The study of the new classes, theory may prove to be extremely vital for future direction in adapting some from of urban design problem solving techniques. It also serves as an evaluation tool for housing design appraisal in the city from Glasgow and similar cities worldwide.
Mr. Virender Narula , Dr. Sandeep Grover,
Volume 26, Issue 1 (3-2015)
Abstract

There has been considerable number of papers published related to Six Sigma applications in manufacturing and service organizations. However, very few studies are done on reviewing the literature of Six Sigma in all the areas including manufacturing, construction, education, financial service, BPOs and healthcare etc. Considering the contribution of Six Sigma in recent time, a more comprehensive review is presented in this paper. The authors have reviewed Six Sigma literature in the way that would help research academicians and practitioners to take a closer look at the growth, development, and applications of this technique. The authors have reviewed various journal papers and suggested different schemes of classification. In addition, certain gap areas are identified that would help researchers in further research.
N. Desai, S. Venkatramana, B.v.d.s. Sekhar,
Volume 31, Issue 3 (9-2020)
Abstract

Today's digital world demands about automated sentiment analysis on visual and text content to significantly displaying people's feelings, opinions and emotions through text, images and videos across popular social networks. Earlier visual sentimental analysis faces many drawbacks like achieve low accuracy and more difficult to understand people opinions due to traditional techniques. Also, another major challenge is a huge number of images generated and uploaded every day across the world. This paper overcomes problems of visual sentiment analysis with the help of deep learning convolution neural network (CNN) and Affective Regions approach to achieve more meaningful sentiment reports with huge accuracy.
Smiljka Miškić, Željko Stević, Ilija Tanackov,
Volume 32, Issue 4 (12-2021)
Abstract

In the field of logistics, there is a daily need for decision making, i.e. the need to solve business problems by selecting an appropriate solution. During the implementation of decision-making processes, it is necessary to find an optimal solution that will best meet the needs of companies. The selection of an optimal solution is crucial for the profitability, cost-effectiveness and long-term development of companies. The decision-making process in logistics is facilitated by applying various tools such as multi-criteria decision-making methods. In this paper, an integrated SWARA (Step-wise Weight Assessment Ratio Analysis) – MARCOS (Measurement Alternatives and Ranking according to Compromise Solution) model was developed and applied in order to classify products. Fifty alternatives, i.e. products were evaluated based on three criteria. The first criterion is the quantity of purchased products, the second criterion is the unit price of products and the third criterion is the annual value of purchase. The SWARA method was applied to determine the significance of the criteria, while the classification of products was performed using the MARCOS method. According to the results of the originally created MCDM model, the products were grouped into three categories A, B, and C. Then, a sensitivity analysis was performed using a model involving the integration of SWARA method and ABC analysis. Using this model, the classification of products into three groups was performed on the basis of the aforementioned criteria, and then a comparative analysis was conducted.
Theodore Alvin Hartanto, Seng Hansun,
Volume 35, Issue 3 (9-2024)
Abstract

One method to diagnose retinal diseases is by using the Optical Coherence Tomography (OCT) scans. Annually, it is estimated that around 30 million OCT scans are performed worldwide. However, the process of analyzing and diagnosing OCT scan results by an ophthalmologist requires a long time so machine learning, especially deep learning, can be utilized to shorten the diagnosis process and speed up the treatment process. In this study, several pre-trained deep learning models are compared, including EfficientNet-B0, ResNet-50V2, Inception-V3, and DenseNet-169. These models will be fine-tuned and trained with a dataset containing OCT scanned images to classify four retinal conditions, namely Choroidal Neovascularization (CNV), Diabetic Macular Edema (DME), Drusen, and Normal. The models that have been trained are then tested to classify the test set and the results are evaluated using a confusion matrix in terms of accuracy, recall, precision, and F1-score. The results show that the model with the best classification results in the batch size of 32 scenario is the ResNet-50V2 model with an accuracy value of 98.24%, precision of 98.25%, recall of 98.24%, and F1-score of 98.24%. While for the batch size of 64, the EfficientNet-B0 model is the model with the best classification results with an accuracy value of 96.59%, precision of 96.84%, recall of 96.59%, and F1-score of 96.59%.


Page 1 from 1