Jannah, Miftahul (2019) Integrasi Data Produk Halal Berdasarkan Jarak Kesamaan Konseptual dan Tekstual Kata. Other thesis, Institut Teknologi Sepuluh Nopember.
Preview |
Text
05211540000153-Undergraduate_Theses.pdf Download (2MB) | Preview |
Abstract
Halal Nutrition Food merupakan produk riset berupa aplikasi yang memberikan pengguna informasi tentang produk makanan halal. Tidak hanya itu, Halal Nutrition Food juga menyediakan informasi nutrisi dan komposisi dari produk tersebut. Namun, data yang ada pada aplikasi ini masih perlu diperkaya dan diolah kembali. Open Food Facts merupakan situs yang menyediakan database mengenai produk makanan (seperti nama produk, komposisi, zat aditif) dimana semua orang di berbagai belahan dunia bisa berkontribusi untuk menambahkan data ke dalamnya maupun menggunakan kembali data yang ada. Pada tugas akhir ini dilakukan pengambilan data dari open food facts kemudian diintegrasikan dengan data pada aplikasi halal nutrition food. Data produk yang akan diintegrasikan diunduh dari situs web open food facts. Data yang ada sangat kotor dan perlu diolah sebelum dilakukan ke proses selanjutnya. Data tersebut diolah pada proses pre-processing data sedemikian rupa sehingga data yang ada dapat diolah ke proses pengukuran kemiripan data. Untuk mengurangi redundansi pada data bahan makanan, penulis melakukan pengukuran kemiripan bahan makanan menggunakan pengukuran kemiripan konseptual dengan Wordnet yang mengukur kemiripan antara dua dataset secara makna kata (sinonim). Selain itu, penulis juga mengukur kemiripan bahan makanan secara tekstual dengan fuzzy string matching, yaitu Levenshtein distance, Jaro-Winkler distance, dan Jaccard distance. Setelah pengukuran kemiripan bahan makanan dilakukan, hasil pengukuran kemiripan diuji untuk mengetahui akurasinya. Dari pengujian tersebut, ditemukan bahwa kombinasi pengukuran kemiripan menggunakan Levenshtein distance (tekstual) dan Wordnet similarity (konseptual) adalah yang paling optimal mengukur kemiripan data bahan makanan sehingga bahan makanan yang diintegrasikan bebas dari redundansi.
=================================================================================================================================
Halal Nutrition Food is a research product in the form of an application that gives users information about halal food products. Not only that, Halal Nutrition Food also provides nutritional information and composition of these products. However, the data contained in this application still needs to be enriched and reprocessed. Open Food Facts is a site that provides a database of food products (such as product names, compositions, additives) where everyone in various parts of the world can contribute to adding data to it or reusing existing data. In this final project, data retrieval from open food facts is then integrated with the data on the halal nutrition food application. Product data to be integrated is downloaded from the open food facts website. The data is very dirty and needs to be processed before going to the next process. The data is processed in the process of pre-processing data in such a way that existing data can be processed into the process of measuring data similarity. To reduce redundancy in food data, measurements of the similarity of food ingredients were carried out using measurement of conceptual similarity with Wordnet which measures the similarity between two datasets in word meaning (synonym). In addition, the authors also measured textual similarity of foodstuffs with fuzzy string matching, namely Levenshtein distance, Jaro-Winkler distance, and Jaccard distance. After measuring the similarity of food ingredients, the results of the similarity measurements were tested to determine their accuracy. From the test, it was found that the combination of measurements of similarity using Levenshtein distance (textual) and Wordnet similarity (conceptual) is the most optimal measure of the similarity of food data so that food ingredients are integrated free from redundancy.
| Item Type: | Thesis (Other) |
|---|---|
| Additional Information: | RSSI 005.74 Jan i-1 2019 |
| Uncontrolled Keywords: | halal, open food facts, wordnet, levenshtein distance, jaro-winkler distance, jaccard distance, integrasi. |
| Subjects: | T Technology > T Technology (General) > T57.5 Data Processing |
| Divisions: | Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Information System > 57201-(S1) Undergraduate Thesis |
| Depositing User: | Miftahul Jannah |
| Date Deposited: | 09 Oct 2024 06:29 |
| Last Modified: | 09 Oct 2024 06:29 |
| URI: | http://repository.its.ac.id/id/eprint/64521 |
Actions (login required)
 |
View Item |