Sharifah Sakinah Syed Ahmad

A conceptual model to empower software requirements conflict detection and resolution with rule-based reasoning

It is seldom technical issues which impede the process of eliciting software requirements. The involvement of multiple stakeholders usually leads to conflicts and therefore the need of conflict detection and resolution effort is crucial. This paper presents a conceptual model to further improve current efforts. Hence, this paper forwards an improved conceptual model to assist the conflict detection and resolution effort which extends the model ability and improves overall performance. The significant of the new model is to empower the automation of conflicts detection and its severity level with rule-based reasoning.

Land cover classification of Landsat 8 satellite data based on Fuzzy Logic approach

The aim of this paper is to propose a method to classify the land covers of a satellite image based on fuzzy rule-based system approach. The study uses bands in Landsat 8 and other indices, such as Normalized Difference Water Index (NDWI), Normalized difference built-up index (NDBI) and Normalized Difference Vegetation Index (NDVI) as input for the fuzzy inference system. The selected three indices represent our main three classes called water, built- up land, and vegetation. The combination of the original multispectral bands and selected indices provide more information about the image. The parameter selection of fuzzy membership is performed by using a supervised method known as ANFIS (Adaptive neuro fuzzy inference system) training. The fuzzy system is tested for the classification on the land cover image that covers Klang Valley area. The results showed that the fuzzy system approach is effective and can be explored and implemented for other areas of Landsat data.

Volunteer selection based on crowdsourcing approach

Voluntary work is important today’s world. There are various versions of the volunteer management system referenced in industry resources. Several organizations have developed volunteer management systems designed to incorporate spontaneous volunteers. However, it can be difficult to find and recruit suitable volunteers for volunteer organizations because the volunteers have many criteria to match with tasks. Also, we still have lacking information on the process of crowdsourcing in volunteering perspective. This paper, we conduct a review of volunteering management systems and crowdsourcing approach. Based on the insights derived from this analysis, we identify some issues for future research. To solve this problem, we designed a framework for the crowdsourcing approach in volunteering system to automate the process of selection volunteers and match with the criteria of volunteers and tasks. Crowdsourcing is one of the best approaches to get more information and faster from the crowd and to be more precise with the requirement from beneficiaries. Fuzzy systems are suitable for such decision-making environments. The implications of the findings for volunteering system are discussed, and future research directions suggested.

PCA and LDA as dimension reduction for individuality of handwriting in writer verification

Principal Component Analysis and Linear Discriminant Analysis are the most popular approach used in statistical data analysis. Both of these approaches are usually implemented as traditional linear technique for Dimension reduction approach. Dimension reduction is useful approach in data analysis application. The concept of dimension reduction will help the process of identifying the most important features in handwritten data which also called as individuality of the handwriting. Where, this individuality will help the verification process in order to verify the handwritten document. The purposed of this paper is to perform both techniques above in writer verification process in order to acquire the individuality of the handwriting. Classification process will be use to evaluate the effectiveness of both approach performance in form of classification accuracy.

A Crowdsourcing Approach for Volunteering System Using Delphi Method

Voluntary work is an important part of virtually every civilization and society. There are several versions of the volunteer management theory referenced in industry resources. Several organizations have developed organization-based systems designed to incorporate spontaneous volunteers. However, it can be difficult to find and recruit suitable candidates for volunteer organizations because the volunteers have many criteria to match with tasks. Also, we still have lacking information on the process of crowdsourcing in volunteering perspective. This paper, we conduct a review of volunteering management systems and crowdsourcing approach. Based on the insights derived from this analysis, we identify some issues for future research. To solve this problem, we designed a framework for the crowdsourcing approach in volunteering system to automate the process of selection volunteers and match with the criteria of volunteers and tasks. Crowdsourcing is one of the best approaches to get more information and faster from the crowd and to be more precise with the requirement from beneficiaries. We are using the Delphi method to criteria from volunteer organization. The implications of the findings for volunteering system are discussed, and future research directions suggested.

Classification of Landsat 8 Satellite Data using Fuzzy c-means

The study aims to classify Landsat 8 satellite data using fuzzy c-means (FCM) method of classification. Fuzzy classification for a land cover is representing to solve mixture classes of pixels. FCM is a method of clustering by assigning the membership degree into one or more cluster. By using both two indices, Normalized Difference Vegetation Index (NDVI) and Normalize Difference Water Index (NDWI) are used as input in FCM clustering to represent the three classes of land cover; vegetation, non-vegetation and water. The classification process is evaluated based on the user, producer, overall accuracy and kappa coefficient.

A Crowdsourcing Approach for Volunteering System

There are several versions of the volunteer management theory referenced in industry resources. Several organizations have developed organization-based systems designed to incorporate spontaneous volunteers. However, it can be difficult to find and recruit suitable candidates for volunteer organizations. Also, we still lack information on the process of crowdsourcing in volunteering perspective. This paper, we conduct a review of volunteering management systems and crowdsourcing approach. Based on the insights derived from this analysis, we identify some issues for future research. To solve this problem, we designed a framework for the crowdsourcing approach in volunteering system to automate the process of selection volunteers and match with the criteria of volunteers and tasks. Crowdsourcing is one of the best approaches to get more information and faster from the crowd and to be more precise with the requirement from beneficiaries. The implications of the findings for volunteering system are discussed, and future research directions suggested.