Claim Missing Document
Check
Articles

Found 3 Documents
Search

Viewer Movie Predictions based on Genres, Actors, and Directors based on Data Mining Using the Eclat Algorithm Deden Muhamad Furqon; Riki Ahmad Maulana; Ahmad Fauzi; Nurul Dwi Cahya; Muhammad Nur Sidiq; Tri Kurnia Sandi
Gunung Djati Conference Series Vol. 3 (2021): Mini Seminar Kelas Data Mining 2020
Publisher : UIN Sunan Gunung Djati Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (483.386 KB)

Abstract

Movies are a place for everyone to find pleasure and entertainment. The film industry is an essential part of the economy in this world. On average, 79% of people in the world enjoy watching movies for their entertainment. Therefore, the film industry has become a huge industry, but it is difficult to predict because the audience's desires are very diverse. Therefore, we created a prediction system for user choices based on the recommendations that will be presented, most likely to be enjoyed by audiences based on genre, actor, and their favorite director, which will function for producers to analyze the market. The method used is data mining using the Eclat algorithm, which has five processes in it. The result is that we get and sort 5043 data by 28 columns and omit some with a support threshold of 0.003 and evaluate the results after evaluating the film data obtained from 1169 lines with a value above 0.003 supporting data to predict user choice recommendations.
Implementation of the ID3 Algorithm for Classification of Car Quality Dimas Adi Putra Pratama; Nisvy Sya`'bana Nugraha; Risnandy Maulana; Adellia Rahmasari; Ahmad Fauzi
Gunung Djati Conference Series Vol. 3 (2021): Mini Seminar Kelas Data Mining 2020
Publisher : UIN Sunan Gunung Djati Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (647.432 KB)

Abstract

Transportation like a car is needed by most of family. Cheap, economical and comfortable family cars have been sold in the market. However, buyers must be careful in finding the car they want. To get the best car, many factors are taken into consideration, such as price, seating, and comfort. Comfort is the main thing in owning a car. There are cars that are expensive but do not have comfort, and vice versa. Therefore, we discuss how the Iterative Dichtomizer Three (ID3) Algorithm can be implemented for the classification process of any car with quality according to price, seat and passenger comfort.
Analysis of Film Budget and Profit using the Bisecting K-Means Algorithm Ahmad Fauzi; Deden Muhamad Furqon; Riki Ahmad Maulana; Nurul Dwi Cahya; Muhammad Nur Sidiq
Gunung Djati Conference Series Vol. 3 (2021): Mini Seminar Kelas Data Mining 2020
Publisher : UIN Sunan Gunung Djati Bandung

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (216.624 KB)

Abstract

As the development of the film industry has become increasingly competitive in the last few decades, the film ecosystem needs to get more attention for stakeholders involved in it to continue to carry out various innovative actions and create more effective creative economy marketing strategies. By utilizing existing datasets, film content producers can build a recommendation system that can support the process of assessing business models and analyzing the concept of creative film products that will be launched in the existing film market so that later planning and design of more profitable film production concepts can be produced. And sustainability in terms of funding (budget) and projected revenue (gross profit). This research is a form of elaboration on the design of a recommendation system using the Bisecting K-Means Algorithm to be able to produce an analysis result in the form of classification of various film products contained in a dataset that has been collected as many as 5048 rows of data by taking 1000 lines of data. As a special data allocation for conducting training on the system. The data that has been allocated will then be carried out by the clustering process by dividing into 4 (four) different clusters, each of which is the result of regression based on predetermined parameters and forming a centroids which is the average value of all cluster nodes built.