Abstrak

Digital forensics starts to show its role and contribution in the society as a solution in disclosure of cybercrime. The essential in digital forensics is chain of custody, which is an attempt to preserve the integrity of digital evidence as well as a procedure for performing documentation chronologically toward evidence. The characteristics of digital evidence have caused the handling chain of custody is becoming more complicated and complex. A number of researchers have contributed to provide solutions for the digital chain custody through a different point of views. This paper gives an overview of the extent to which the problem and challenges are faced in the digital chain of custody issue as well as the scope of researches that can be done to contribute in the issue of the digital chain of custody. General TermsDigital Forensics

The rapid development of the internet and social media and a large amount of text data has become an important research subject in obtaining information from the text data. In recent years, there has been an increase in research on sentiment analysis in the review text to determine the polarity of opinion on social media. However, there are still few studies that apply the deep learning method, namely Long Short-Term Memory for sentiment analysis in Indonesian texts.This study aims to classify Indonesian novel novels based on positive, neutral and negative sentiments using the Long Short-Term Memory (LSTM) method. The dataset used is a review of Indonesian language novels taken from the goodreads.com site. In the testing process, the LSTM method will be compared with the Naïve Bayes method based on the calculation of the values of accuracy, precision, recall, f-measure.Based on the test results show that the Long Short-Term Memory method has better accuracy results than the Naïve Bayes method with an accuracy value of 72.85%, 73% precision, 72% recall, and 72% f-measure compared to the results of the Naïve Bayes method accuracy with accuracy value of 67.88%, precision 69%, recall 68%, and f-measure 68%.

Abstrack -In an effort to anticipate the occurrence of errors in the selection of prospective borrowers while improving the quality of customer service, finance companies need decision making tools that simplify and speed up the process of predicting prospective borrowers who are able to pay off credit. The study discusses the application design process in constructing decision tree using C4.5 algorithm and utilizing a group of training data of motorcycle financing debtor, then interpreted in the form of decision rule as a reference in estimating potential loan repayment of debtor. The test results through 5 categories of tests performed in the process of the resulting tree required an average time of 112 seconds with the fastest time obtained in the first test category with the amount of data 3000 records worth 9 seconds. While in the process of generate rule it takes an average time of 1.78 seconds with the fastest time obtained in the first test category with the amount of data 3000 records worth 1.23 seconds. Comparison of the amount of data in each test category affects the value of execution time, the more data make longer the process of generating trees and rules. In the data accuracy test obtained the average percentage of data accuracy value of 51.2% with the highest gain in the first test category with total data 3000 records worth 54%.Intisari -. Dalam upaya mengantisipasi terjadinya kesalahan dalam pemilihan calon debitur sekaligus meningkatkan kualitas layanan konsumen, perusahaan pembiayaan membutuhkan alat bantu pengambilan keputusan sehingga mempermudah dan mempercepat proses prediksi calon debitur yang mampu melunasi kredit. Penelitian membahas proses rancangn bangun aplikasi dalam membangun pohon keputusan menggunakan algoritma C4.5 dan memanfaatkan sekelompok data latih debitur pembiayaan kendaraan sepeda motor, kemudian diinterpretasikan dalam bentuk aturan keputusan sebagai acuan dalam memperkirakan potensi pelunasan kredit calon debitur. Hasil pengujian melalui 5 kategori uji yang dilakukan dalam proses generate tree dibutuhkan rata-rata waktu 112 detik dengan perolehan waktu tercepat pada kategori uji pertama dengan jumlah data 3000 record senilai 9 detik. Sedangkan dalam proses generate rules dibutuhkan rata-rata waktu 1,78 detik dengan perolehan waktu tercepat pada kategori uji pertama dengan dengan jumlah data 3000 record senilai 1,23 detik. Perbandingan jumlah data disetiap kategori uji mempengaruhi nilai execution time, makin banyak datanya maka semakin lama untuk proses generate tree dan rules. Pada pengujian akurasi data diperoleh prosentase rata-rata nilai akurasi data 51,2% dengan perolehan tertinggi pada kategori uji pertama dengan total data 3000 record senilai 54%.Kata kunci: debitur, kredit, C4.5, pohon keputusan

The Police as law enforcers who authorize in terms of social protection are expected to do both the prevention and investigation efforts also the settlement of criminal cases that occurred in the society. This research can help police to identify the main actor faster and leads to solving crime-cases. The use of overall centrality is very helpful in determining the main actors from other centrality measures. The purpose of this research is to identify the central actor of crimes done by several people. Semantic Social Network Analysis is used to perform central actor identification using five centrality measurements, such as degree centrality, betweenness centrality, closeness centrality, eigenvector centrality, and overall centrality. As for the relationship between actors, this research used social relation such as friendship, colleague, family, date or lover, and acquaintances. The relationship between actors is measured by first four centrality measures then accumulated by overall centrality to determine the main actor. The result showed 80.39% accuracy from 102 criminal cases collected with at least 3 actors involved in each case.

Abstrak Kata kunci-algoritma semut, masalah pengambilan dan pengiriman, penghematan tertinggi, urun daya, konsolidasi perjalanan Abstract Common practice in crowdsourced delivery services is through direct delivery. That is by dispatching direct trip to a driver nearby the origin location. The total distance can be reduced through multiple pickup and delivery by increasing the number of requests in a trip. The research implements exact algorithm to solve the consolidation problem with up to 3 requests in a trip. Greedy heuristic is performed to construct initial route based on highest savings. The result is then optimized using Ant Colony Optimization (ACO

AbstrakSebagai  daerah yang rawan bencana dari tsunami, dan abrasi gelombang laut, dibutuhkan suatu lokasi yang layak untuk dapat dijadikan suatu daerah pemukiman untuk mendukung perkembangan perekonomian, sarana dan prasarana, serta sosial ekonomi di pemerintah kota Lhokseumawe. Oleh sebab itu, diperlukan penentuan kelayakan lokasi pemukiman yang layak untuk direkomendasi menjadi sebuah pemukiman yang akan dibangun. Sehingga perlu dilakukan evaluasi terhadap areal fisik pemukiman, sarana dan prasarana, dan sosial-ekonomi yang akan direkomendasi menjadi suatu lokasi pemukiman. Masukan dari masyarakat sangat dibutuhkan untuk mengetahui masalah lokasi yang akan direkomendasi menjadi sebuah pemukiman. Hal ini dilakukan agar pihak pengawas mengetahui dan dapat mengantisipasi masalah-masalah yang terjadi pada lokasi tersebut. Penerapan sistem pendukung keputusan kelompok atau Group Decision Support System (GDSS) yang dibuat menggunakan  metode VIKOR (Vlse Kriterijumska Optimizacija  Kompromisno Resenje in Serbia) dapat membantu para pengambil keputusan dalam melakukan perankingan nilai masing-masing alternatife solusi, entropy sebagai pembobotannya dari setiap kriteria.. Sementara metode Copeland score, sebagai salah satu metode voting yang tekniknya berdasarkan pengurangan frekwensi kemenangan dengan frekwensi kekalahan dari perbandingan berpasangan untuk melakukan voting terhadap solusi yang akan di ambil para decision maker. Kata kunci—  GDSS, Vikor, Entropy, Copeland Score, lokasi pemukiman.  AbstrakAs a disaster-prone areas of the tsunami, ocean waves and abrasion, it takes a decent location to be used as a residential area to support economic development, infrastructure, social and economic development of Regional Goverment of Lhokseumawe. Therefore, it is necessary to have feasibility determination of viable residential locations as recommendation to be a settlement. So it is necessary to evaluate the physical area of settlements, infrastructure, economic and social development that will be recommended to be a residential location. Public suggestion is needed to determine the problem locations to be recommended to a settlement. The application of group decision support systems or Group Decision Support System (GDSS) are made using the VIKOR method (Vlse Kriterijumska Optimizacija Kompromisno Resenje in Serbia) that is able to assist decision makers in the ranking value to each solution alternatives, entropy wih weigt for criterias, the method of Copeland score, as one method of voting is the technique by reducing the frequency of victory with the defeat of the comparison frequency pairs can be used to vote for a solution that will take the decision makers. Keyword— GDSS, Vikor, Copeland Score, Entropy, location determination.

One approach that is often used in forecasting is artificial neural networks (ANN), but ANNs have problems in determining the initial weight value between connections, a long time to reach convergent, and minimum local problems.Deep Belief Network (DBN) model is proposed to improve ANN's ability to forecast exchange rates. DBN is composed of a Restricted Boltzmann Machine (RBM) stack. The DBN structure is optimally determined through experiments. The Adam method is applied to accelerate learning in DBN because it is able to achieve good results quickly compared to other stochastic optimization methods such as Stochastic Gradient Descent (SGD) by maintaining the level of learning for each parameter.Tests are carried out on USD / IDR daily exchange rate data and four evaluation criteria are adopted to evaluate the performance of the proposed method. The DBN-Adam model produces RMSE 59.0635004, MAE 46.406739, MAPE 0.34652. DBN-Adam is also able to reach the point of convergence quickly, where this result is able to outperform the DBN-SGD model.