Claim Missing Document
Check
Articles

Found 3 Documents
Search
Journal : INFOKUM

Implementation of Greedy Algorithm for Profit and Cost Analysis of Swallow's Nest Processing Dirty to Finished Products Efendi Efendi; Daniel Ryan Hamonangan Sitompul; Stiven Hamonangan Sinurat; Ruben Ruben; Andreas Situmorang; Dennis Jusuf Ziegel; Julfikar Rahmad; Evta Indra
INFOKUM Vol. 10 No. 02 (2022): Juni, Data Mining, Image Processing, and artificial intelligence
Publisher : Sean Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (425.252 KB)

Abstract

Swallow's nest is made from the saliva of swallows, especially species of swallows of the genus Collocalia. Swallow's nest is used traditionally to improve health so it is widely consumed by the community. Swallow nest products are difficult to produce, causing the product to be expensive. This study aims to analyze the costs and benefits of swallow nest production. The analysis uses the Greedy algorithm, which is looking for solutions to each stage of production. The principle of Greedy's algorithm is "take what you can get now". There are 6 processes in the production of swiftlet nests, namely sorting raw materials, cleaning, drying, printing, in process control (IPC) and packaging. In the sorting and cleaning process, employees in the medium and medium to light nest categories were combined. The total costs incurred in the sorting process are reduced by 14% and the costs incurred in the cleaning process are reduced by 8%. The process of drying dense and medium hair nests takes the same time so that they are carried out simultaneously and the required cost is reduced by 11% to Rp 675,000. The stages of printing the original and super types of nests are combined because they have.
SENTIMENT ANALYSIS COMPARE LINEAR REGRESSION AND DECISION TREE REGRESSION ALGORITHM TO DETERMINE FILM RATING ACCURACY Rivaldo Sitanggang; Daniel Ryan Hamonangan Sitompul; Stiven Hamonangan Sinurat; Ruben, Andreas Situmorang; Denis Jusuf Ziegel; Julfikar Rahmad; Evta Indra
INFOKUM Vol. 10 No. 02 (2022): Juni, Data Mining, Image Processing, and artificial intelligence
Publisher : Sean Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (567.849 KB)

Abstract

Rating assessment in a film is the most important thing because it describes the satisfaction of film lovers with the films they have watched. With technological advances like now, we can easily find out the rating of a film by using a platform to accommodate the audience's review results, namely the Internet Movie Database (Imdb). The Machune Learning model that has been created can determine whether the film we watch is good based on ratings and reviews from moviegoers who share their experiences in watching similar films. Based on the results of the analysis of the two algorithms Linear Regression and Dicision Tree Regression, the best accuracy results from the Decision Tree Regression algorithm are 95.47%
APPLICATION OF DATA MINING TO PREDICATE STOCK PRICE USING LONG SHORT TERM MEMORY METHOD Sonia Novel Lase; Yenny Yenny; Owen Owen; Mardi Turnip; Evta Indra
INFOKUM Vol. 10 No. 02 (2022): Juni, Data Mining, Image Processing, and artificial intelligence
Publisher : Sean Institute

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (218.315 KB)

Abstract

Investing some of our wealth to invest in stocks is highly recommended considering the fluctuating nature of stock prices, meaning that stock prices can go up and down at any time depending on the conditions and phenomena that occur on the stock market. Stock investment includes having a high risk of loss but also by taking that risk it is also possible to get high profits (High Risk High Return). Shares are proof of ownership of company value or proof of equity interest. Shareholders are also entitled to receive dividends (profit sharing) according to the number of shares they own. This study aims to make it easier for everyone who wants to invest in Google and Tesla stocks and implement the long short term memory method for stock price prediction. This data mining research resulted in a Root Mean Square Error (RMSE) value of 1.80%, which means the prediction results are very accurate with real data and the average difference between real stock price data and predicted data is $3 -$15.