Sistem Rekomendasi Pemilihan Produk UMKM Berbasis Hybrid Recommendation

Abstract

Smeska is one of the entrepreneurship programs belonging to the Solo Techno Incubator which is structurally related to Solo Technopark and is under the auspices of the Regional Research and Development Agency for the City of Surakarta. This program provides entrepreneurship training for non-digital startups or MSMEs with the hope that the output of the participants will be able to carry out end-to-end processes (production, branding, marketing, digitization & packaging) as well as setting the team's focus on the area of business achievement. With the special segmentation of MSME players, until the third year this program was running, it was still not far from the conventional scale even though digitization had become the point in the training. One solution that can accommodate this digitalization is to design a MSME product recommendation system that makes it easier for the client side, in this case buyers, to find what products they need. The purpose of this study is to make a Hybrid Recommendation model for the MSME Product Selection Recommendation System. The system development method used is Extreme Programming (XP) which consists of stages namely planning, design, implementation, as well as testing and integration. The Hybrid Recommendation modelling design used in this study is Pipelined Hybridization where the first recommender in this system is Content Based with naïve bayes techniques which will be input to the second recommender using Knowledge Based modelling with Case Based techniques. Modelling for this MSME product selection recommendation system can provide filtering of the search for conformity product items along with 5 choices of product search attributes, namely brand, price, material, variant and size. Based on the results of the Hybrid Recommendation modelling with 20 sample data, Adzkira chips get the highest similarity value of 0.99000 from the search input for MSME product types of chips. It is hoped that the results of this research can make a significant contribution to developing a recommendation system for selecting MSME products, as well as strengthening digitalization and business transformation efforts for Smeska program participants.