C4.5 ALGORITHM OF DECISION TREE

Le but du présent rapport est de présenter deux algorithmes utilisés dans le cadre du datamining. Dans un premier temps nous présenterons l’algorithme classique qu’est ID3, nous aborderons cet algorithme qui est un algorithme d’apprentissage supervisé, puis forts de cette étude nous aborderons plus en détails C4.5 Celui-ci étant une extension naturelle de l’algorithme ID3. Les deux algorithmes ont été présentés par J. Ross Quinlan dans les années 90 pour découvrir des modèles de classification de la donnée, aussi appelé arbres de décisions, à partir de données. Et on va faire une comparaison entre ces deux algorithme et d’autres.

Untuk menunjang kinerja, sebuah perusahaan harus memiliki sumber daya manusia yang dapat diandalkan. Dalam memperoleh sumber daya manusia atau yang biasa disebut pegawai perusahaan, maka perusahaan melakukan proses seleksi atau rekrutmen untuk penerimaan pegawai. Dalam rangka mendapatkan SDM terbaik, maka proses seleksi harus dilaksanakan dengan baik dan benar. Proses seleksi SDM memerlukan cara yang terorganisir dan akurat agar menghasilkan SDM yang bermutu dan dapat mendukung kesuksesan organisasi. SDM merupakan sebuah investasi bagi organisasi maupun perusahaan untuk dapat tumbuh dan berkembang sesuai dengan visi dan misi perusahaan.
PT. Telkom merupakan perusahaan BUMN yang bergerak dibidang jasa layanan telekomunikasi. Tentunya banyak pelamar dari seluruh Indonesia yang ingin bekerja di PT. Telkom. Oleh karena itu diperlukan objektivitas dalam pengambilan keputusan. Pengambilan keputusan yang dilakukan oleh manusia masih banyak yang bersifat subjektif. Belum lagi resiko nepotisme yang tidak dapat dihindari. Bila dibiarkan dalam jangka panjang, hal tersebut dapat mempengaruhi kinerja organisasi. Tujuan seleksi pada dasarnya untuk mendapat orang yang tepat untuk jabatan tertentu agar dapat bekerja secara optimal dalam waktu yang lama. Proses seleksi penerimaan pegawai di perusahaan cukup panjang sehingga membutuhkan waktu yang cukup lama dan biaya yang tidak sedikit dan tentunya itu semua untuk mendapatkan calon pegawai yang benar-benar sesuai dengan kriteria yang tepat.
Tujuan penelitian ini adalah membuat klasifikasi penerimaan pegawai dengan menggunakan Decision Tree oleh Algoritma C4.5. Hasil klasifikasi ini dapat digunakan untuk mengambil keputusan dalam proses penerimaan pegawai.

Bu tezin amacı kalp ameliyatı sırasında ya da kalp ameliyatı geçirdikten kısa bir süre sonra hastaya ait hayati riskin sınıflandırmaya dayalı makine öğrenmesi teknikleri kullanılarak belirlenebilmesidir. Çalışmada kullanılan veri seti Acıbadem Maslak Hastanesi'nden temin edilmiştir. Literatürde kalp ameliyatı sırasında ya da kalp ameliyatı geçirdikten kısa bir süre sonra hastanın ölüm riskinin tahmini için kullanılan EuroSCORE (The European System for Cardiac Operative Risk Evaluation)'a ait risk faktörleri kullanılmıştır. Veri setindeki gözlemlerde EuroSCORE'daki gibi 30 günlük takip bilgisi bulunmadığından, öncelikle hastaların Standart EuroSCORE puanları hesaplanmıştır. Sonrasında bu risk grupları sınıf etiketi olarak kullanılmış ve tahminler gerçekleştirilmiştir. Naive Bayes Sınıflandırıcı, k-En Yakın Komşu Algoritması, Logistik Regresyon Analizi, ID3 ve C4.5 Karar Ağacı Algoritmaları kullanılarak farklı modeller oluşturulmuştur. Modellerin performansları karşılaştırılmıştır. Veri analizleri R dilinde yazılan kodlarla gerçekleştirilmiştir. R kodlarını geliştirme aracı olarak RStudio kullanılmıştır. Logistik Regresyon Analizinden elde edilen modeller, Shiny (shinyapps.io) aracılığı ile web üzerinden kamuya açık hale getirilmiştir. En iyi performansı gösteren C4.5 karar ağacı modeli için ise başka bir Shiny uygulaması geliştirilmiştir.

Karyawan merupakan faktor terpenting dalam suatu perusahaan, terutama dalam memberikan pelayanan pada konsumen. Kriteria karyawan sangat menentukan apakah karyawan tersebut berhak untuk memperoleh kenaikan gaji atau tidak. Melihat kondisi tersebut, penilaian atau evaluasi terhadap kinerja karyawan dirasa perlu untuk dlakukan. Penilaian dilakukan dengan menggunakan Sistem Pendukung Keputusan Menentukan Status Kenaikan Gaji Karyawan Menggunakan Metode C4.5. Dalam pembuatan sistem pendukung keputusan ini, ada beberapa tahap yang dilakukan diantaranya : pengumpulan data, menghitung jumlah data, jumlah keputusan gaji naik, jumlah keputusan gaji tetap dan entropy semua kasus yang dibagi per masing-masing atribut, lalu dilanjutkan dengan perhitungan gain untuk masing-masing atribut. Dengan adanya sistem ini, diharapkan dapat membantu pihak manajemen dalam proses penilaian kinerja karyawan untuk keperluan kenaikan gaji karyawan.

Kata Kunci : C4.5 , Karyawan, Keputusan.

Educational Data Mining (EDM) is a study methodology and an application of data mining techniques related to student’s data from academic database. Like other domain, educational domain also produce vast amount of studying data. To enhance the quality of education system student performance analysis plays an important role for decision support. This paper elaborates a study on various Educational data mining technique and how they could be used to educational system to analysis student performance. It helps the teacher to identify students who need special attention and allow the teacher to provide appropriate guidance. This paper showcases the importance of Classification based data mining algorithms in the field of education system and also presents some promising future success.

Keywords: , Classification, Decision Tree, ID3, C4.5.

Credit is a belief that one is given to a person or other entity which is concerned in the future will fulfill all the obligations previously agreed. some of the problems that often occur in the crediting institutions eg consumer arrears were previously deemed worthy of receiving credit, the breakdown of credit status. The emergence of the problem as a lack of consideration or the stability of credit analysis in determining credit worthiness when consumers apply for crediting. Therefore it is necessary to do credit analysis so as to determine the feasibility of a credit crunch, through credit analysis results, it can be seen whether the customer is feasible or not. Some of the existing problems of classification method is used to predict credit is by using models classification algorithm C4.5 After testing with the model showed that 90.99% and AUC value of 0.911 to the level of diagnosis Excellent Classification. and it can be concluded that the application of the C4.5 classification algorithm capable of generating patterns of creditworthiness with pinpoint accuracy and a good diagnosis

Obesitas merupakan suatu masalah kesahatan yang efek negatif dengan konsekuensi panjang bagi pengidapnya, keluarga, maupun orang lain. Data Riskesdas menunjukkan bahwa sebanyak 21,7% penduduk dewasa Indonesia yang berusia lebih dari 18 tahun mengidap obesitas. Obesitas juga merupakan akar dari penyakit-penyakit berbahaya lainnya seperti penyakit jantung koroner (PJK) dan cenderung menjadi diabetogenik yang akan mengkontaminasi penyakit-penyakit berbahaya seperti hiperlipidemia, penyakit hati dan kantong empedu, osteoartristis, kanker, dan penyakit saluran pernapasan. Selain menyebabkan masalah fisiologis obesitas juga menyebabkan masalah emosional dan psikologis seperti berkurangnya kepercayaan diri karena penampilan fisik ’kurang menarik’. Data Mining Data mining merupakan suatu metode atau teknik yang digunakan untuk memecahkan suatu permasalah data menggunakan implementasi-implementasi untuk menemukan solusi data. Model Decision Tree Algortima C4.5 merupakan salah satu algoritma yang digunakan proses penggalian data (data mining) yang sangat kuat dan terkenal. Pada penelitian ini, kami mengusulkan sebuah evaluasi model Decision Tree Algortima C4.5 untuk proses klasifikasi sebagai solusi prediktif kemungkinan terjadinya obesitas. Kemudian Model ini dievaluasi dan divalidasi menggunakan WEKA 3.7.4 guna menghitung nilai permanci model yang digunakan.

Abstrak Kemiskinan merupakan salah satu masalah yang dialami oleh beberapa Negara berkembang, termasuk indonesia. Banyak cara yang dilakukan untuk menanggulangi kemiskinan, diantaranya dengan program bantuan sosial untuk rakyat miskin. Bentuk bantuan sosial yang diberikan oleh pemerintah disesuaikan dengan tingkat kemiskinan yang ada disuatu wilayah sehingga pemberian bantuan sosial tersebut tidak salah sasaran. Pada penelitian kali ini kami menggunakan BDT (Basis Data Terpadu) yang dikeluarkan oleh TNP2K dalam menentukan klasifikasi tingkat kemiskinan. Adapun metode yang digunakan dalam penelitian ini adalah Naïve Bayes Clasifier (NBC) dan Algoritma C4.5 yang keduanya merupakan metode pada teknik klasifikasi data mining. Pegujian akan dilakukan dengan menggunakan 14 atribut. Hasil dari proses klasifikasi diperoleh bahwa metode C4.5 memiliki tingkat akurasi 3% lebih baik jika dibandingkan dengan metode Naïve Bayes. Indonesia merupakan salah satu Negara berkembang di asia khususnya asia tenggara. Salah satu masalah yang sering dihadapi oleh Negara berkembang adalah kemiskinan. Berdasarkan data yang dikeluarkan oleh Badan Pusat Statistik, angka kemiskinan di Indonesia pada tahun 1999 mencapai 47.97 juta jiwa. Pada tahun 2011 jumlah penduduk miskin menjadi 30.02 juta jiwa. Badan Pusat Statistik melakukan pendataan kependudukan khususnya masalah kemiskinan setiap 3 tahun sekali. Proses pendataan dilakukan dengan cara door to door langsung menuju rumah tangga sasaran. Bagi pemerintah Indonesia masalah kemiskinan merupakan masalah lama yang belum dan sulit untuk diselesaikan [8]. Pemerintah sendiri telah melakukan beberapa upaya dalam melakukan pengentasan kemiskinan diantaranya melalui program bantuan sosial diantaranya Bantuan langsung tunai (BLT), Program Keluarga Harapan (PKH) dll. Salah satu kesulitan yang terkadang dihadapi oleh pemerintah adalah proses pembagian bantuan sosial yang tidak merata dan tepat sasaran. Kendala tersebut bisa terjadi karena faktor teknis maupun nonteknis. Basis Data Terpadu yang dikeluarkan oleh TNP2K merupakan salah satu dasar yang digunakan dalam penentuan klasifikasi tingkat kemiskinan tersebut. Di dalamnya terdapat setidaknya 14 atribut yang menentukan tingkat kemiskinan diantaranya tingkat pendidikan, jenis lapangan usaha, kedudukan dalam pekerjaan, status tempat tinggal, jenis atap, jenis dinding, jenis lantai, air minum yang digunakan, penerangan yang digunakan, bahan bakar memasak, fasilitas buang air besar, fasilitas tempat pembuangan akhir, kepesertaan KB, serta jumlah anggota rumah tangga dalam satu rumah. Salah satu pemodelan yang bisa digunakan untuk klasifikasi tingkat kemiskinan tersebut adalah dengan menggunakan data mining. Dalam penelitian kali ini penulis mencoba membandingkan model klasifikasi yang dibentuk oleh teknik data mining antara algoritma decision tree C4.5 dan Naïve Bayes Clasifier (NBC). Metode ini di uji cobakan dengan menggunakan jumlah data

The use of mobile phones has exploded over the past years, abundantly through the introduction of smartphones and the rapidly expanding use of mobile data. This has resulted in a spiraling problem of ensuring quality of service for users of mobile networks. Hence, mobile carriers and service providers need to determine how to prioritise expansion decisions and optimise network faults to ensure customer satisfaction and optimal network performance. To assist in that decision-making process, this research employs data mining classification of different Key Performance Indicator datasets to develop a monitoring scheme for mobile networks as a means of identifying the causes of network malfunctions. Then, the data are clustered to observe the characteristics of the technical areas with the use of k-means clustering. The data output is further trained with decision tree classification algorithms. The end result was that this method of network optimisation allowed for significantly improved fault detection performance.

ان تعدين البيانات فعالية الحصول على المعرفة لتحقيق هدف أساس وهو اكتشاف الحقائق الخفية (Hidden Facts) التي تتضمنها قواعد البيانات وذلك من خلال استخدام تقنيات متعددة تشتمل على الذكاء الاصطناعي، التحليلات الاحصائية، تقنيات و نمذجة البيانات ... الخ. فأن عملية تعدين البيانات تولد نماذج وعلاقات واضحة في البيانات والتي تساعد على توقع النتائج في المستقبل. وقد ظهرت العديد من الخوارزميات التي في هذا المجال، وترتب عليها مقارنة بين هذه الخوارزميات لاختبار الخوارزمية المناسبة في الحصول على نتائج أفضل. وقد هدف البحث الى استخدام المصنف C4.5 وربطها مع الشبكة العصبية نوع Back –  Propagation (BP) وذلك لتكوين نموذج تصنيف يحمل خواص الطرفين، فضلا عن مقارنة النتائج المستحصلة مع نتائج التصنيف باستخدام الحزمة البرمجية الجاهزة Minitab . وتوصل البحث الى ان المعادلات الخاصة بالمصنف C4.5 كانت أكفأ في الاداء وخاصة بعد ربطها بالشبكة العصبية BP لازالة التناقض والتشويش الموجود في البيانات، كما عززت النتائج من افضلية استخدام لغات البرمجة مقارنة بنتائج التطبيق الجاهزة.

Intrusion detection system dealing with the huge amount of data that include repeated irrelevant cause slow process of testing, training and higher learning resource consumption as well as the vulnerability of the detection rate. Data mining techniques are being applied in the construction of intrusion detection systems to protect computing resources against unauthorized access. In this paper, we have done a comparative study on machine learning tools using WEKA and Rapid Miner with two classifier algorithms C4.5 and Decision Stump for Network Intrusion Detection to measure the accuracy, sensitivity and precision. The results of the experiments using the KDD' 99 attack dataset and select seven features, The results show the best tools Rapid Miner for the accuracy and precision, while the best algorithms is C4.5.

Depuis  l’avènement  des  disciplines  de  l’intelligence  artificielle  et  de  l’apprentissage automatique  des  méthodes et  des  techniques  ne  cessent  d’émerger,  on  notera  que celles-ci sont soit issu du courant symbolique (particulièrement au début) ou du statistique. Dans ce contexte, les arbres de décision se situent au confluent de ces deux courants, ils sont basés sur une approche purement statistique pour la construction  mais une fois construit ils permettent une bonne lecture de leur structure et une interprétation aisée de leurs résultats. Un  arbre  de  décision  est  un  classifieur  simple, graphique  et  un  outil  supervisé  d’aide  à  la décision.  Il  est lisible,  rapide  en  apprentissage  et  en  exécution,  toutes  ces caractéristiques expliquent sa popularité. Toutefois, le principal reproche que l’on  fait à  ces  classifieurs  est leur instabilité. Dans ce papier, nous nous intéressons à l’un des éléments qui causent cette instabilité ; il s’agit du critère de choix de la variable de segmentation lors de la construction de  l’arbre. Nous  effectuons  alors  une  comparaison  entre  les principaux  algorithmes  de construction  pour  arbres  de  décision,  pour évaluer  l’impact  de  ce  choix  aussi  bien  sur  la structure de l’arbre que sur ses performances.

In cancer identification, extrapolation of cervical cancer in patients shows a vital role. To save people from the cancer field of cancer detection, machine learning can play a big role in saving lives. In this paper, to make the detection process a portion faster and accurate machine learning techniques such as Decision Stump, C4.5 and Averaged One Dependence Estimators (AODE) for novel NCBI cervical cancer data set are made. Classification and Regression Tree (CART) is a simple decision tree algorithm that is used to create a decision tree of a given set. Here, a top-down Greedy search is used in order to check each attribute at every tree node. For building a Decision Stump algorithm, a decision tree which consists of nodes and an arc that connects nodes with the Entropy concept is used. The extension of the basic Decision Stump algorithm is the C4.5 algorithm on selecting the optimal split it recursively visits each decision node. The process gets continued until there is no further split is possible. In this way, the prediction is possible for the given data set. Bayesian methods are those that explicitly apply Bayes' Theorem for problems such as classification algorithms acts as a quick method for the creation of a statistical predictive model. AODE are based on the Bayesian theorem which is commonly used to solve prediction problems for ease usage in the medical field. In this research Decision Stump, C4.5, and AODE are implemented with help of the training set. The basic designs are used to predict whether a feminine is having cervical cancer or not.

C 4.5 is the best known classification algorithm used
to generate decision trees for continuous and discrete
attributes. Although its extension C5.0 is further
developed, it is still used in variety of applications. It
works on information gain ratio of given attributes.
The limitations of C4.5 is its information entropy, it
gives poor results for larger distinct attributes. In
this work we proposed a modified approach to
overcome the draw backs of C4.5 based on
information gain and gain ratio.
Key words: - C4.5 Algorithm, Information gain,
Entropy, Gain Ratio.

This study aims to optimize the selection of raw water source options at the location of Pamsimas III program. This analysis will affect the condition of the facilities that will function properly. The data mining method used in this study is a classification model with the Naive Bayes algorithm using the Rapidminer application tool. The data processed is SIM Pamsimas III data with the object of research on the new village drinking water facilities for the 2017-2019 Pamsimas program. This research analyzes the prediction of the condition of drinking water facilities based on the option of raw water sources. So this research can helps to determine the level of potential facilities that will function properly based on the specified raw water source options. Based on the research conducted, predictive analysis using Naive Bayes has an accuracy rate of 85.05%. So that the selection of the raw water source option can predict the condition of the drinking water facilities being built will function properly.

Classification is widely used technique in the data mining domain, where scalability and efficiency are the immediate problems in classification algorithms for large databases. We suggest improvements to the existing C4.5 decision tree algorithm. In this paper attribute oriented induction (AOI) and relevance analysis are incorporated with concept hierarchy's knowledge and HeightBalancePriority algorithm for construction of decision tree along with Multi level mining. The assignment of priorities to attributes is done by evaluating information entropy, at different levels of abstraction for building decision tree using HeightBalancePriority algorithm. Modified DMQL queries are used to understand and explore the shortcomings of the decision trees generated by C4.5 classifier for education dataset and the results are compared with the proposed approach.

Breast cancer is one of the leading cancers for women in developed countries including India. It is the second most common cause of cancer death in women. The high incidence of breast cancer in women has increased significantly in the last years. In this paper we have discussed various data mining approaches that have been utilized for breast cancer diagnosis and prognosis. Breast Cancer Diagnosis is distinguishing of benign from malignant breast lumps and Breast Cancer Prognosis predicts when Breast Cancer is to recur in patients that have had their cancers excised. This study paper summarizes various review and technical articles on breast cancer diagnosis and prognosis also we focus on current research being carried out using the data mining techniques to enhance the breast cancer diagnosis and prognosis.

Data Mining refers to using a variety of techniques to classify the propose of information or decision making knowledge in the database .The healthcare industry collects huge amounts of healthcare records which, unfortunately, are not " mined " to discover hidden information for effective decision making .The data from medical history has been found as diverse data and it seems that the various forms of data should be interpret to envisage the heart disease of a patient. Various techniques in Data Mining have been applied to predict the patients of heart disease .Using data mining classification techniques such as Decision Tree Algorithm, Naïve Bayes, and Neural Network. etc., the patient risk level is classified .Accuracy of the risk level is high when using more number of attributes. Results show that each technique has its unique strength in realizing the objectives of the defined mining goals .It can predict the likelihood of patients getting a heart disease using medical profiles such as age, sex, blood pressure and blood sugar.

This paper describes the application of two decision tree algorithms, Logistic Regression Trees with Unbiased Selection (LOTUS) and Classification Rule with Unbiased Interaction Selection and Estimation (CRUISE), to identify prevailing distress types that affect pavement maintenance decision making. The maintenance history of the selected sections within Interstate highway I-90 in the State of Wisconsin is modeled to build decision trees. The historical field distress survey data include the types, quantities, severities, and locations of distresses. The primary objective of this paper is to present the decision tree algorithms that may be incorporated into the decision making process and help state highway agencies make rational decisions during the course of pavement management. The paper discusses the model development process and prevailing distress identification procedures employing the decision tree algorithms. A case study is conducted on another Interstate highway, I-43, with a view to furthering the analysis of the results from the two algorithms. These analysis models and the results will provide the state highway agencies and practitioners with a means to identify prevailing distress types in a specific location and facilitate making decisions on future maintenance practices.

Classifying and recognizing Telugu characters (aksharas) is a challenging task
because of the variations in the script and the large number of characters. The
complexity of the shape is a result of structural compositions involving vowels
(V), consonants (C), consonants with vowel modifiers (CV) and consonant
clusters (CCV). This paper presents a novel classification strategy for
classifying aksharas with the CV structure. This is achieved by constructing
two decision trees — one for consonants and the other for vowel modifiers —
by hybridizing the moment invariants and C4.5 algorithm. The results,
although preliminary and limited in scale, illustrate the potential strengths of
the approach. Perhaps, the uniqueness and the strength of the approach lies in
the divide-and-conquer strategy where the two orthographically orthogonal
decision trees each recognizing only a few tens of classes may in combination
potentially classify hundreds of classes

The data mining techniques have the ability to discover hidden patterns or correlation among the objects in the medical data. There are many areas that adapt data mining techniques, namely marketing, stock, health care sector and so on. In the health care industry produces gigantic quantities of data that clutches complex information relating to the sick person and their medical conditions. The data mining has an infinite potential to make use of healthcare data more effectually and efficiently to predict various kinds of disease. The present-time healthcare industry heart ailment is a term that assigns to an enormous number of health care circumstances related to heart. These medical circumstances relate to the unexpected health circumstance that straight control the cardiac. In this paper we are using a ROCK algorithm because it uses Jaccard coefficient on the contrary using the distance measures to find the similarity between the data or documents to classify the clusters and the contrivance for classifying the clusters based on the similarity measure shall be used over a given set of data. Afterward, C4.5 algorithm is used as the training algorithm to show the rank of a cardiac ailment with the decision tree. The C4.5 can be referred as the statistic classifier as well as this algorithm uses avail radio for feature selection and to build the decision tree. The C4.5 algorithm is widely used because of its expeditious classification and high exactitude. Lastly, the cardiac ailment database is clustered using the K-means clustering, which will alienate the data convenient to cardiac sickness from the database.

Abstrak Keterlambatan merupakan tindakan menyimpang yang menyalahi aturan atau tata tertib yang ada di sekolah baik tertulis maupun tidak tertulis. Kedisiplinan siswa datang ke sekolah menjadi yang pertama yang dilihat, beberapa faktor umum yang terjadi keterlambatan bisa terjadi antara lain: jarak ke sekolah, jam bangun, jam berangkat, kondisi perjalanan, dan kendaraan yang dipakai. Dalam penelitian ini, penulis menggunakan algoritma Klasifikasi dengan metode C4.5 dan metode Forward selection .Sampel yang digunakan adalah data kuesioner siswa-siswa kelas VIII (delapan) di Sekolah Menengah Pertama Negeri 271 sebanyak 270 siswa. Penggunaan data training, ditentukan atribut-atribut tertentu untuk membentuk model classifer. Hasil dari penelitian ini adalah hasil akurasi metode C4.5 sebesar 60,74% dengan hasil tree menunjukkan kemacetan merupakan faktor keterlambatan sekolah dan hasil akurasi 65.93% untuk Forward selection dan mendapatkan 3 atribut terbaik. Abstrack Delay is a deviant act that violates the rules or orderliness in the school both written and unwritten. The discipline of students coming to school is the first to be seen, some common factors that delays can occur, example: distance to school, waking hours, departure hours, travel conditions, and vehicles used. In this research, the authors used the Classification algorithm with the C4.5 method and the Forward selection method. The sample used was questionnaire data for class VIII (eight) students at 271 Junior High School as many as 270 students. Using training data, certain attributes are determined to form a classifer model. The results is the accuracy of the C4.5 method of 60.74% with the results of the tree indicating congestion is a factor of school delay and the result of accuracy is 65.93% to Forward selection and get the 3 best attributes. PENDAHULUAN Keterlambatan merupakan tindakan menyimpang yang menyalahi aturan atau tata tertib yang ada di sekolah baik tertulis maupun tidak tertulis. Mengajarkan kedisiplinan pada anak usia sekolah atau siswa merupakan hal yang sangat penting untuk membuat siswa mengerti akan peraturan yang ada di sekolah. Akan tetapi, fakta di lapangan menunjukkan bahwa masih banyak siswa yang sering datang terlambat ke sekolah. Keterlambatan datang ke sekolah tersebut bukan berarti tanpa sebab (Sulistiyono 2018), berbagai macam alasan diungkapkan para siswa yang sering terlambat. Pelanggaran terhadap tata tertib memang banyak dijumpai di sekolah (Akhmad Rizkon 2019) yang umumnya dilakukan oleh para siswa. Pelanggaran adalah tindakan menyalahi aturan yang dilakukan oleh seseorang dengan sengaja (Insyiroh and Naqiyah 2017) Ada beberapa macam penyebab siswa datang terlambat ke sekolah, waktu tidur malam yang telat, menonton film atau TV terlalu malam, kebiasaan bangun pagi telat, jarak dari rumah ke sekolah (Pramono et al. 2018) serta kondisi perjalanan dan kendaraan yang dipakai (Amirulloh and Taufiqurrochman 2017) Algoritma yang digunakan adalah metode C4.5 dan jenis seleksi fitur yang digunakan adalah Forward selection yaitu pengelolaan variabel yang telah dipilih melalui seleksi fitur dan peningkatan akurasi prediksi (Saleh 2017). Tujuan dari penelitian untuk membantu mengatas permasalah siswa terlambat datang ke sekolah, dapat memahami faktor-faktor ketelambatan siswa datang ke sekolah, dan membantu pihak sekolah dalam menanggulangi

We consider the re-identification of users of on-line social networks when they participate in several different on-line social networks, potentially using several different accounts. The re-identification of users serves several purposes: (i) commercial use so as to avoid redundant mailing to the same user; (ii) enhancement of the information available about these users by unifying information from different sources; (iii) consolidation of accounts by on-line social network providers; (iv) identification of potentially malicious users and/or bots. We highlight that all this should occur within the bounds of the data protection and privacy laws as well as the users' expectations on such matters to avoid backlash. In this paper, we explore this situation first by a formalization using the SAN model to conceptually structure information as a graph, which includes user and attribute type nodes. This formalization enables us to reason on two issues. First, how to identify that two or more user-accounts belong to the same user. Second, what gains in predictability are obtained after re-identification. For the first issue, we show that a set-difference approach is remarkably effective. For the second issue we explore the impact of re-identification on the predictability by two different machine learning algorithms: C4.5 (decision tree induction) and SVM-SMO (Support Vector Machine with SMO kernel). Our results show that as predictability improves, in some cases different SAN metrics emerge as predictors.

Abstrak-Penelitian ini memberikan satu model penentuan kelayakan pemberian fasilitas pembiayaan. Kendala yang sering ditemukan dalam proses pembiayaan di sebuah perusahaan yaitu sulitnya menentukan nasabah mana yang termasuk kedalam kelompok lancar dan tidak lancar. Oleh karena itu diperlukan suatu aplikasi yang membantu perusahaan dalam proses pengklasifikasian nasabah. Untuk menentukan calon nasabah mana yang termasuk kelompok lancar dan tidak lancar. Data dalam penelitian ini diambil dari perusahaan dalam jumlah yang cukup banyak proses ini bisa disebut sebagai data mining. Dalam penelitian ini untuk menganalisa data digunakan teknik decision tree dengan memanfaatkan algoritma C4.5. Dalam perancangan aplikasi menggunakan diagram konek, DFD, dan penyajian aplikasi menggunakan Borland Delphi 7 serta database Xampp. Hasil yang di capai adalah sebuah Aplikasi Algoritma C4.5 Untuk Pengklasifikasian Calon Nasabah yang Implementasinya dapat membantu bagian Data Enty dalam penyeleksian dan penerimaan

Today, enormous amount of data is collected in medical databases. These databases may contain valuable information encapsulated in nontrivial relationships among symptoms and diagnoses. Extracting such dependencies from historical data is much easier to done by using medical systems. Such knowledge can be used in future medical decision making. In this paper, a new algorithm based on C4.5 to mind data for medince applications proposed and then it is evaluated against two datasets and C4.5 algorithm in terms of accuracy.

ABSTRACT Power system islanding contingency has always been one of the most severe wide area system failures. The recent Hurricane Gustav landed in west coast in 2008 was one exact case to illuminate the importance and severity of this kind of contingencies. Longtime the researches of islanding situations focus on studying the coherence groups of the generations under an islanding contingency. In this paper, a new method in detecting power system islanding contingencies using both the system's topological structure and real-time system dynamic state variables is presented and tested. An islanding severity index concept is given for ranking the severity of the islanding cases.