PENGGUNAAN DATA MINING DALAM HIT RATE IMPORTASI JALUR MERAH DENGAN MODEL DECISION TREE

ABSTRACT: Customs and Excise faces a big challenge to be able to increase the hit rate of red line imports by 40% in accordance with the Blueprint for the 2014-2025 Ministry of Finance Institutional Transformation Program and international benchmarks. Through a qualitative study, this study aims to determine the use of data mining that is applied to the risk engine based on import data, people's experiences, and research results of customs institutions of other countries. The data mining method used is CRISP-DM, classification method, and decision tree model, using data imported from the red line KPU BC Type A Tanjung Priok for the period September – December 2019 and January 2020. The results show that the use of data mining can increase the hit rate of red line importation. The most relevant attribute in classifying data is the sending country which is categorized as a root node, while the import duty tariff attribute does not provide information on data classification. This research is expected to provide a new perspective for the KPU BC Type A Tanjung Priok in an effort to improve the risk engine targeting and risk engine routing of Customs and Excise. Keywords: CRISP-DM, data mining, decision tree, hit rate, the red line import. ABSTRAK: Bea dan Cukai menghadapi tantangan besar untuk dapat meningkatkan capaian hit rate importasi jalur merah sebesar 40% sesuai dengan Cetak Biru Program Transformasi Kelembagaan Kementerian Keuangan Tahun 2014 – 2025 dan benchmark internasional. Melalui studi kualitatif, penelitian ini bertujuan untuk mengetahui penggunaan data mining yang diterapkan dalam risk engine berdasarkan data importasi, pengalaman orang, dan data hasil penelitian institusi kepabeanan negara lain. Metode data mining yang digunakan adalah CRISP-DM, metode klasifikasi, dan model decision tree, dengan menggunakan data importasi jalur merah Kantor Pelayanan Utama (KPU) Bea dan Cukai (BC) Tipe A Tanjung Priok periode September – Desember 2019 dan Januari 2020. Hasil penelitian menunjukkan bahwa penggunaan data mining dapat meningkatkan capaian hit rate importasi jalur merah. Atribut yang paling relevan dalam mengklasifikasikan data adalah negara pengirim yang dikategorikan sebagai root node (akar), sedangkan atribut tarif bea masuk tidak memberikan informasi dalam klasifikasi data. Penelitian ini diharapkan dapat memberikan pandangan baru bagi KPU BC Tipe A Tanjung Priok dalam upaya perbaikan risk engine targeting dan risk engine penjaluran Bea dan Cukai. Kata Kunci: CRISP-DM, data mining, decision tree, hit rate, importasi jalur merah.

Spatial decision tree model for garlic land suitability evaluation

Predicting land and weather characteristics as indicators of land suitability is very important in increasing effectiveness in food production. This study aims to evaluate the suitability of garlic land using spatial decision tree algorithm. The algorithm is the improvement of the conventional decision tree algorithm in which spatial join relation is included to grow up spatial decision tree. The spatial dataset consists of a target layer that represents garlic land suitability and ten explanatory layers that represent land and weather characteristics in the study areas of Magetan and Solok district, Indonesia. This study generated the best spatial decision trees for each study area. On Magetan dataset, the best model has 33 rules with 94.34% accuracy and relief variable as the root node, whereas on Solok dataset, the best model has 66 rules with 60.29% accuracy and soil texture variable as the root node.

Implementasi Iterative Dichotomiser 3 (ID3) Untuk Penentuan Kelayakan Pemberian Kredit Pada PT.BPR Ploso Saranaartha Jombang

Pada PT. BPR Ploso Saranaartha Jombang terdapat beberapa permasalahan yang kerap muncul mengenai pemberian kredit pada debitur. Saat ini bank tersebut memberikan kredit kepada nasabahnya masih dengan cara seleksi debitur, belum ada prosedur secara sistematis dalam menentukan apakah suatu nasabah tersebut layak diberikan kredit atau tidak. Hal inilah yang mengakibatkan banyaknya kredit macet yang dapat merugikan pihak bank.. Algoritma Iterative Dichotomiser 3 (ID3) merupakan salah satu algoritma yang dapat digunakan untuk menyelesaikan kasus ini. Dalam penyelesaiiannya nanti ID3 akan melakukan proses preprocessing data terlebih dahulu yang bertujuan untuk membuang data-data yang tidak penting guna mendapatkan data yang benar-benar dibutuhkan. Setelah itu ID3 akan membentuk sebuah pohon keputusan berdasarkan rules yang dihasilkan. Setiap root node dalam sebuah pohon keputusan terbentuk berdasarkan nilai gain terbesar dari tiap-tiap input atribut. Dalam perhitungan algoritma ini diperlukan sebuah dataset yang cukup untuk digunakan proses training. Dataset yang digunakan untuk proses training ini berjumlah 300 record data yang terdiri dari 272 data dengan kolektibilitas lancar dan 28 data dengan kolektibilitas macet. Terdapat pula data yang akan digunakan untuk proses testing berjumlah 20 record data nasabah baru yang terdiri dari 8 data dengan kolektibilitas macet dan 12 data dengan kolektibilitas lancar. Dalam uji coba yang telah dilakukan terhadap dataset menghasilkan 10 rules. Setelah dilakukan proses testing atau pengujian terhadap data testing dihasilkan output dengan tingkat keakurasian 88,51%. Artinya dari 300 record data yang telah ditraining dapat meng-cover 19 data dari 20 record data testing.

Implementasi Algoritma Decision Tree (J.48) untuk Memprediksi Resiko Kredit pada BMT

Kredit merupakan hal utama pada lembaga keuangan yang berpengaruh pada pertumbuhan dan perkembangan lembaga tersebut. Lemahnya pengawasan dan manajemen dalam proses pemberian kredit kepada nasabah dapat menyebabkan tingginya kredit macet. Masalah ini terjadi pada salah satu lembaga keuangan yang memberikan kredit pada nasabah yaitu BMT X. Data tahun 2019 menunjukkan terdapat 600 ajuan kredit multiguna. Dari jumlah tersebut, hanya sekitar 76% menunjukkan kolektabilitas yang baik. Kondisi kolektabilitas kredit yang tidak maksimal menyebabkan BMT X harus mengeluarkan biaya lebih untuk mengumpulkan angsuran yang harus dibayarkan oleh debitur secara langsung. Kredit macet ini menimbulkan kerugian pada lembaga keuangan yang bersangkutan. Untuk itu, dalam memberikan kredit, BMT X harus cerdas menilai kelayakan nasabah. Tujuan penelitian ini adalah menyusun rancangan kebijakan BMT X guna meminimasir kesalahan prediksi nasabah dengan kategori kredit macet. Teknik yang digunakan pada penelitian ini yaitu data mining klasifikasi dengan algoritma J.48. Untuk mengukur efektivitas suatu atribut dalam mengklasifikasikan kumpulan sampel data, harus dipilih atribut yang memiliki information gain terbesar yang akan diletakkan pada root node. Penelitian ini menghasilkan enam rule dengan tingkat akurasi sebesar 80,2% sehingga dapat digunakan pihak BMT X untuk menggali informasi kelayakan nasabah untuk mendapatkan kredit. Kata kunci: Algoritma J.48, data mining, pohon keputusan, resiko kredit

Exploring the impact of unstable terminals on branch support values in paleontological data

Recent studies have acknowledged the many benefits of including fossils in phylogenetic inference (e.g., reducing long-branch attraction). However, unstable taxa are known to be problematic, as they can reduce either the resolution of the strict consensus or branch support. In this study, we evaluate whether unstable taxa that reduce consensus resolution affect support values, and the extent of such impact, under equal and extended implied weighting. Two sets of analyses were conducted across 30 morphological datasets to evaluate complementary aspects. The first focused on the analytical conditions incrementing the terminal instability, while the second assessed whether pruning wildcards improves support. Changes in support were compared with the “number of nodes collapsed by unstable terminals,” their “distance to the root,” the “proportion of missing data in a dataset,” and the “proportion of sampled characters.” Our results indicate that the proportion of missing entries distributed among closely related taxa (for a given character) might be as detrimental for stability as those distributed among characters (for a given terminal). Unstable terminals that (1) collapse few nodes or (2) are closely located to the root node have more influence on the estimated support values. Weighting characters according to their extra steps while assuming that missing entries contribute to their homoplasy reduced the instability of wildcards. Our results suggest that increasing character sampling and using extended implied weighting decreases the impact of wildcard terminals. This study provides insights for designing future research dealing with unstable terminals, a typical problem of paleontological data.

Finding the best-fit background function for whole-powder-pattern fitting using LASSO combined with tree search

A new linear function for modelling the background in whole-powder-pattern fitting has been derived by applying LASSO (least absolute shrinkage and selection operator) and the technique of tree search. The background function (BGF) consists of terms b n L(2θ/180)−n/2 and b n H(1 − 2θ/180)−n/2 for the low- and high-angle sides, respectively. Some variable parameters of the BGF should be fixed at zero while others should be varied in order to find the best fit for a given data set without inducing overfitting. The LASSO algorithm can automatically select the variables in linear regression analysis. However, it finds the best-fit BGF with a set of adjustable parameters for a given data set while it derives a different set of parameters for a different data set. Thus, LASSO derives multiple solutions depending on the data set used. By regarding the individual solutions from LASSO as nodes of trees, tree structures were constructed from these solutions. The root node has the maximum number of adjustable parameters, P. P decreases with descending levels of the tree one by one, and leaf nodes have just one parameter. By evaluating individual solutions (nodes) by their χ2 index, the best-fit single path from a root node to a leaf node was found. The present BGF can be used simply by varying P in the range 1–10. The BGF thus derived as a final single solution was incorporated into computer programs for Pawley-based whole-powder-pattern decomposition and Rietveld refinement, and the performance of the BGF was tested in comparison with the polynomials currently widely used as the BGF. The present BGF has been demonstrated to be stable and to give an excellent fit, comparable to polynomials but with a smaller number of adjustable parameters and without introducing undulation into the calculated background curve. Basic algorithms used in statistics and machine learning have been demonstrated to be useful in developing an analytical model in X-ray crystallography.

Optimizing RPL performance based on the selection of best route between child and root node using E-MHOF method

IETF has proposed the routing protocol for low power and lossy networks (RPL) for IOT as view as light weight routing protocol. In RPL, the objective function (OF) is used to select the best route between child and root node. Several researches have been conducted in order to, enhance OF according to number parameters such as number of hops, remaining energy and expected number of transmissions (ETX), without a consideration to other challenges such as congestion node problem and latency. So, to overcome these challenges a new technique called “Enhance-Minimum Rank with Hysteresis Objective Function (MHOF)” is proposed in this paper, to select the ideal path between the child and root node. The technique is consisted of three layers: parent selection layer in which parent is selected based on three parameters (ETX, RSSI and nodes’ residual energy), path selection layer in which the best route is chosen according to the minimum of (average ETX value) and maximum of (average remaining energy value) of all nodes in the selected route. The last layer is child node minimization, which utilized to solve the congestion node energy problem by using two parameters (RSSI reference and threshold value). The proposed method has been implemented and evaluated by using Cooja simulator software. The simulation results have shown that selected path with E-MHOF is increased the network lifetime and reduced latency in comparison with MHOF.

Geminate Acquisition and Representation by Ammani Arabic-Speaking Children

This study investigates the acquisition and representation of geminate plosives and geminate liquids in the speech of Ammani Arabic (AA) children aged from 2;6 (year; month) to 5 years, at six months intervals. Although gemination is perceived properly by AA children at an early stage, in which they produce words including geminates significantly longer than words with singletons, the acoustic measurements indicate that the children’s phonetic/acoustic representations start to be noticeable (in comparison with adults) at the age-stage of 3;7-4, and much closer to that of the adults by the age-stage 4;7-5. In terms of phonological representation, it is found that gemination is implemented as a main strategy by AA children for word-medial clusters at syllable boundaries. Interestingly, if the medial cluster is not pronounced faithfully, it will be pronounced as a geminate consonant where the second member of the cluster compensates for the deleted consonant to rescue the moraic weight and the segmental length that would be achieved by producing the target of two distinct consonantal gestures. This strategy offers an intriguing piece of evidence for the two-root node composite modal which combines moraic representation of geminates (by preserving weight to the syllable) and prosodic representation (by preserving segmental length to the geminate consonant).

The deep(er) roots of Eukaryotes and Akaryotes

Background: Locating the root node of the “tree of life” (ToL) is one of the hardest problems in phylogenetics, given the time depth. The root-node, or the universal common ancestor (UCA), groups descendants into organismal clades/domains. Two notable variants of the two-domains ToL (2D-ToL) have gained support recently. One 2D-ToL posits that eukaryotes (organisms with nuclei) and akaryotes (organisms without nuclei) are sister clades that diverged from the UCA, and that Asgard archaea are sister to other archaea. The other 2D-ToL proposes that eukaryotes emerged from within archaea and places Asgard archaea as sister to eukaryotes. Williams et al. ( Nature Ecol. Evol. 4: 138–147; 2020) re-evaluated the data and methods that support the competing two-domains proposals and concluded that eukaryotes are the closest relatives of Asgard archaea. Critique: The poor resolution of the archaea in their analysis, despite employing amino acid alignments from thousands of proteins and the best-fitting substitution models, contradicts their conclusions. We argue that they overlooked important aspects of estimating evolutionary relatedness and assessing phylogenetic signal in empirical data. Which 2D-ToL is better supported depends on which kind of molecular features are better for resolving common ancestors at the roots of clades – protein-domains or their component amino acids. We focus on phylogenetic character reconstructions necessary to describe the UCA or its closest descendants in the absence of reliable fossils. Clarifications: It is well known that different character types present different perspectives on evolutionary history that relate to different phylogenetic depths. We show that protein structural-domains support more reliable phylogenetic reconstructions of deep-diverging clades in the ToL. Accordingly, Eukaryotes and Akaryotes are better supported clades in a 2D-ToL.