#TITLE_ALTERNATIVE#

• Main Author: NICOLOSI WASKITA (NIM: 18213041), INDRA
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/22575
• Institution: Institut Teknologi Bandung

One of human's developmental task during early adulthood is selecting a life partner. <br /> <br /> <br /> <br /> <br /> Developmental Process Theory states that life partner are chosen through series of <br /> <br /> <br /> <br /> <br /> gradual filtering processes that focuses on one’s family background and personal <br /> <br /> <br /> <br /> <br /> characteristics. These filtering processes are difficult because it involves many criteria <br /> <br /> <br /> <br /> <br /> of consideration. One example of a technology that is able to facilitate life partner <br /> <br /> <br /> <br /> <br /> filtering processes is Online Dating Sites (ODS). <br /> <br /> <br /> <br /> <br /> There are two common problems found in most ODS. The first problem is the lack of <br /> <br /> <br /> <br /> <br /> consideration of implicit user preferences in the provision of recommendation by the <br /> <br /> <br /> <br /> <br /> system. Users often do not have a full understanding of their own preferences. <br /> <br /> <br /> <br /> <br /> Therefore, in addition to explicit preferences, ODS should also base its <br /> <br /> <br /> <br /> <br /> recommendation on implicit preferences. The second problem is the absence of <br /> <br /> <br /> <br /> <br /> matching criteria weighting feature. Most ODS apply the same weight for each criterion <br /> <br /> <br /> <br /> <br /> to all users. Applying same weight for each user is actually a problem because each <br /> <br /> <br /> <br /> <br /> user has different pattern of prioritizing matching criteria. Both of those problems cause <br /> <br /> <br /> <br /> <br /> the user to not get an accurate life partner recommendation. <br /> <br /> <br /> <br /> <br /> The purpose of this research is to build a prototype of Decision Support System (DSS) <br /> <br /> <br /> <br /> <br /> in a form of ODS that can address the two common problems stated before. The <br /> <br /> <br /> <br /> <br /> recommendation algorithm used by DSS is reciprocal recommendation algorithm, an <br /> <br /> <br /> <br /> <br /> algorithm that apply reciprocal liking concept in giving recommendation. This <br /> <br /> <br /> <br /> <br /> algorithm calculates the value of compatibility between users with two methods namely <br /> <br /> <br /> <br /> <br /> content-based and collaborative filter. The collaborative filter method allows the ODS <br /> <br /> <br /> <br /> <br /> to calculate value of compatibility between users based on implicit preferences obtained <br /> <br /> <br /> <br /> <br /> from historical data of interactions between users. The SPK prototype is equipped with <br /> <br /> <br /> <br /> <br /> a criteria weighting feature based on the Analytical Hierarchy Process (AHP) method. <br /> <br /> <br /> <br /> <br /> AHP facilitates users to indirectly weight matching criterion through a pairwise <br /> <br /> <br /> <br /> <br /> comparison matrix. The prototype of DSS is integrated with a social media analytic tool <br /> <br /> <br /> <br /> <br /> so that DSS can provide additional information for the user. This tool will provide <br /> <br /> <br /> <br /> <br /> information about the attitudes and characteristics of users, tendency of users in <br /> <br /> <br /> <br /> <br /> misusing addictive substances, and users involvement in terrorism. The prototype test <br /> <br /> <br /> <br /> <br /> results show that the prototype has been able to meet the DSS requirements in support <br /> <br /> <br /> <br /> <br /> of the life partner selection process.