Poster based on Dreuw & Deselaer's Poster LaTeX Template (http://www-i6.informatik.rwth-aachen.de/~dreuw/latexbeamerposter.php) and downloaded from http://www.LaTeXTemplates.com.
\begin
Discover why over 20 million people worldwide trust Overleaf with their work.
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% Dreuw & Deselaer's Poster
% LaTeX Template
% Version 1.0 (11/04/13)
%
% Created by:
% Philippe Dreuw and Thomas Deselaers
% http://www-i6.informatik.rwth-aachen.de/~dreuw/latexbeamerposter.php
%
% This template has been downloaded from:
% http://www.LaTeXTemplates.com
%
% License:
% CC BY-NC-SA 3.0 (http://creativecommons.org/licenses/by-nc-sa/3.0/)
%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%----------------------------------------------------------------------------------------
% PACKAGES AND OTHER DOCUMENT CONFIGURATIONS
%----------------------------------------------------------------------------------------
\documentclass[final,hyperref={pdfpagelabels=false}]{beamer}
\usepackage{multirow}
\usepackage[orientation=portrait,size=a0,scale=1.4]{beamerposter} % Use the beamerposter package for laying out the poster with a portrait orientation and an a0 paper size
\usepackage{xcolor}
\usetheme{I6pd2} % Use the I6pd2 theme suplied with this template
%\usepackage{extsizes}
\usepackage[english]{babel} % English language/hyphenation
\usepackage{amsmath,amsthm,amssymb,latexsym} % For including math equations, theorems, symbols, etc
%\usepackage{times}\usefonttheme{professionalfonts} % Uncomment to use Times as the main font
%\usefonttheme[onlymath]{serif} % Uncomment to use a Serif font within math environments
\boldmath % Use bold for everything within the math environment
\usepackage{booktabs} % Top and bottom rules for tables
\graphicspath{{figures/}} % Location of the graphics files
\usecaptiontemplate{\small\structure{\insertcaptionname~\insertcaptionnumber: }\insertcaption} % A fix for figure numbering
%----------------------------------------------------------------------------------------
% TITLE SECTION
%----------------------------------------------------------------------------------------
\title{\huge Marketplace Seller Recommender with User-Based Multi Criteria Decision Making} % Poster title
\author{Murein Miksa Mardhia (0519108901), Dwi Normawati (0504088601) \\
murein.miksa@tif.uad.ac.id , dwi.normawati@tif.uad.ac.id} % Author(s)
\institute{Informatics Engineering Department - Ahmad Dahlan University} % Institution(s)
%----------------------------------------------------------------------------------------
% FOOTER TEXT
%----------------------------------------------------------------------------------------
\newcommand{\leftfoot}{bit.do/poster-PDP093} % Left footer text
\newcommand{\rightfoot}{PDP-093/SKPP/III/2018} % Right footer text
%----------------------------------------------------------------------------------------
\begin{document}
\addtobeamertemplate{block end}{}{\vspace*{2ex}} % White space under blocks
\begin{frame}[t] % The whole poster is enclosed in one beamer frame
\begin{columns}[t] % The whole poster consists of two major columns, each of which can be subdivided further with another \begin{columns} block - the [t] argument aligns each column's content to the top
\begin{column}{.02\textwidth}\end{column} % Empty spacer column
\begin{column}{.465\textwidth} % The first column
%----------------------------------------------------------------------------------------
% INTRODUCTION
%----------------------------------------------------------------------------------------
\begin{block}{Introduction - Background}
\begin{itemize}
\item
This research develops a recommender system for seller selection of several online marketplace in Indonesia. Since users are provided with too many merchants with various offers regarding a product, it then creates such difficulty and longer time for them to decide which merchant they should choose to get the efficient effort but optimum results.
\item
Case studies in several of the most popular online shops were adopted: Tokopedia, Shopee and Bukalapak, where respondents make online transaction the most. Criteria stated are product price, seller location, seller reputation, the number of sold products and expedition support.
\item This study applied a method of Fuzzy Simple Additive Weighting to normalize by using weight authorized from user preference.
\item The experiment applied a user-based method of testing due to each preference and method to place the rankings. At the end, a merchant with the highest point placed as the top rank and displayed as the recommendation.
\end{itemize}
%\begin {itemize}
%\item How to accommodate need and give continuous improvement toward both sides, teachers and KM system?
\end{block}
%----------------------------------------------------------------------------------------
% METHODS
%----------------------------------------------------------------------------------------
\begin{block}{Methods}
\begin{itemize}
\item \textbf{Data Acquisition }
\begin{itemize}
\item This study employs data from Tokopedia, Shopee and Bukalapak online shops about products, sellers and attributes chosen from respondents’ user experience. There are 50 samples of data of products and 120 rank combination orders of respondent data taken from each marketplace.
There are five criteria defined as the weighted attributes:
\begin{enumerate}
\item Products price, identified by the nominal in Rupiah
currency.
\item Distance of seller location to the user’s location,
identified by the information of city or district from the merchant’s page.
\item Sellers reputation, identified by how many star(s)
collected by merchant through users’ feedback.
\item The number of products sold by seller, identified by how many product(s) a merchant had been sold.
\item Number of delivery supports, identified by how many expedition company are supporting the users’ destination area.
\end{enumerate}
\end{itemize}
%\begin{figure}
%\includegraphics [width=0.4\linewidth]{figures/paper-Figure_1.JPG}
%\caption{Framework Method of Seller Recommendation in Marketplace with SAW}
%\end{figure}
\bigskip
\item \textbf {Fuzzy - Simple Additive Weighting}
\begin{itemize}
\item This framework presents a suitable explanation for knowledge management system development. Its main components are stakeholders, processes (business process and knowledge processes), knowledge strategies, infrastructures and results.
\begin{figure}
\includegraphics[width=0.5\linewidth]{figures/paper-Figure_2.JPG}
\caption{Workflow of Fuzzy-SAW Methodology}
\end{figure}
\end{itemize}
\begin{itemize}
\item The weighting part is done by giving a value of a scale of 0-1. After that, in the matrix mapping stage, the data of sellers is sold that sell similar products along with the value of the attributes of each alternative matrix and attributes.
\item The second stage is the stage of normalization of the matrix. Each alternative seller/merchant candidate calculated the normalization value per attribute, adjusted for the type of attribute.
\end{itemize}
\begin{table}[h!]
\caption{RAW MATRIX OF ALTERNATIVES}
\begin{center}
\begin{tabular}{||c| c c c c c||}
\hline
Alternative & C1 & C2 & C3 & C4 & C5 \\ [0.5ex]
\hline\hline
A1 & 9000 & 713 & 4.6 & 701 & 1 \\
\hline
A2 & 29000 & 394 & 3.9 & 0 & 1 \\
\hline
A3 & 12000 & 370 & 4.6 & 646 & 3 \\
\hline
A4 & 12000 & 403 & 4.7 & 26 & 2 \\
\hline
A5 & 11000 & 258 & 4.5 & 342 & 3 \\
\hline
A6 & 29200 & 390 & 4 & 0 & 1 \\
\hline
A7 & 10282 & 208 & 4.6 & 71 & 3 \\
\hline
A8 & 9300 & 399 & 4.5 & 304 & 2 \\
\hline
A9 & 7900 & 713 & 4.5 & 2030 & 1 \\
\hline
A10 & 13000 & 238 & 4.6 & 391 & 1 \\ [1ex]
\hline
\end{tabular}
\end{center}
\end{table}
\end{itemize}
\end{block}
%------------------------------------------------
\end{column} % End of the first column
\begin{column}{.03\textwidth}\end{column} % Empty spacer column
\begin{column}{.465\textwidth} % The second column
%------------------------------------------------
%---
%----------------------------------------------------------------------------------------
% RESULTS
%----------------------------------------------------------------------------------------
\begin{block}{Results: Implementation and Accuracy Evaluation}
\begin {itemize}
\item After accomplishing data pre-processing and implementation, Fig. 2 presents the interface of the personal recommender system.
\begin{figure}
\includegraphics[width=0.5\linewidth]{figures/paper-Figure_3.JPG}
\caption{Recommender System Prototype Interface}
\end{figure}
\begin{itemize}
\item Table 1 illustrates the normalized value matrix toward a Laundry Basket product at Shopee. There are at least ten sellers who sell the product with the same visual (type, brand and size) which will be processed by the Fuzzy-SAW method.
\item The normalized result table is then multiplied by the weighting number. Out of the 120 types of ranking methods, we can take an example of the 1-2-3-4-5 ranking sequence for price attributes-location-reputation-sold product-support expeditions. Three consecutive sellers who get the highest score are A7-A9-A5.
\end{itemize}
\begin{figure}
\includegraphics[width=0.5\linewidth]{figures/paper-Figure_4.JPG}
\caption{Alternative Matrix Calculation Results with 1Weight Matrix}
\end{figure}
\begin{itemize}
\item This study employs an accuracy of the rank result calculation method. The accuracy is calculated based on the recommendations given by the system to data compared to the data given by respondents which is divided into data training to find the most optimum weight; and data testing in ratio of 4:1.
\item In the data testing, a random sample of products with various combination of attributes rank order is applied. Accuracy value of each product search result is defined if a target seller is found as the first recommendation result.
\item Table 2 shows a sample of data testing that compares seller target to output from system. From the table, the accuracy of the correct amount of data can be calculated compared to the number of data testing.
\item Attribute rank show the user preference respectively for price, location, number of sold product, seller reputation, and the number of expedition provider support. The accuracy generated by Fuzzy-SAW in this personal recommender system is 75\%.
\end{itemize}
\begin{table}[h!]
\scriptsize
\caption{SAMPLE OF DATA TESTING}
\begin{center}
\begin{tabular}{ |c|c|c|c| }
\hline
Product & Attribute Rank & Seller Target & Seller Result \\
\hline
\multirow{4}{4em}{Aceh Arabica Coffee}
& 1-5-2-3-4 & Kopi Tubruk Indonesia & Kopi Tubruk Indonesia \\
& 1-2-3-4-5 & Q House of Coffee & Kopi Tubruk Indonesia \\
& 2-1-3-4-5 & Boenboen Coffee & Boenboen Coffee \\
& 3-1-2-4-5 & Kopi Tubruk Indonesia & Kopi Tubruk Indonesia \\ \hline
\multirow{3}{4em}{Zara Floral}
& 1-3-4-2-5 & Kimi Shop & Kimi Shop \\
& 2-4-1-5-3 & Kimi Shop & Value Bags \\
& 3-1-5-4-2 & Kimi Shop & Kimi Shop \\
\hline
\multirow{3}{4em}{Simbadda Music Player}
& 4-1-5-2-3 & Travarillo & Travarillo \\
& 1-2-3-4-5 & IT Shop Online & Simbadda Official \\
& 2-1-5-4-3 & IT Shop Online & IT Shop Online \\
\hline
\end{tabular}
\end{center}
\end{table}
\end{itemize}
\end{block}
\begin{block}{Conclusions}
\begin{itemize}
\item From the results of accuracy and data from participants, the price attribute becomes the attribute that respondents were being considered the most, since it originally belongs to a product, while the other attributes belong to each merchant.
\item Some criteria ranking combinations produce the same sellers who are always ranked in the top-4 recommendations, one of the trigger factors because the values in almost all attributes are the most optimum number of each attribute type.
\end{itemize}
\end{block}
\end{column} % End of the second column
\begin{column}{.015\textwidth}\end{column} % Empty spacer column
\end{columns} % End of all the columns in the poster
%----------------------------------------------------------------------------------------
%----------------------------------------------------------------------------------------
% CONCLUSION
%----------------------------------------------------------------------------------------
%----------------------------------------------------------------------------------------
\begin{center}
\color {white}
\textbf{
This research is funded by Research Grant Program of The Ministry of Research and Higher Education of Indonesia in academic year 2017/2018. }
\bigskip
\end{center}
\end{frame} % End of the enclosing frame
\end{document}