RapidMiner

In this paper the authors proposed different Multilayer Perceptron Models (MLP) of artificial neural networks (ANN) suitable for visual merchandising in Global Distribution (GDO) applications involving supermarket product facing. The models are related to the prediction of different attributes concerning mainly shelf product allocation applying times series forecasting approach. The study highlights the range validity of the sales prediction by analysing different products allocated on a testing shelf. The paper shows the correct procedures able to analyse most guaranteed results, by describing how test and train datasets can be processed. The prediction results are useful in order to design monthly a planogram by taking into account the shelf allocations, the general sales trend, and the promotion activities. The preliminary correlation analysis provided an innovative key reading of the predicted outputs. The testing has been performed by Weka and RapidMiner tools able to predict by MLP ANN each attribute of the experimental dataset. Finally it is formulated an innovative hybrid model which combines Weka prediction outputs as input of the MLP ANN RapidMiner algorithm. This implementation allows to use an artificial testing dataset useful when experimental datasets are composed by few data, thus accelerating the self-learning process of the model. The proposed study is developed within a framework of an industry project.

Research on the performance of workers based on the determination of standard working hours for absences conducted by workers in a certain period. In disciplinary supervision, workers are expected to be able to provide the best performance in the implementation of work in accordance with predetermined working hours. The measurement of the level of discipline of admission hours for placement workers is carried out every working day, continuously and continuously. Attendance monitoring already uses online attendance by using data downloaded from the online attendance provider as the main data. In addition, data collection is done by filtering employee absentee data and supporting information on the categories that cause mismatches in meeting work schedules. Mobilization of workers according to location and working hours has been regulated in company regulations allowing the placement of workers in accordance with the residence so as not to affect the desired work results the company is still within reasonable limits and can be increased. The assessment of this study as a progression factor inhibiting the company in achieving company targets. From the results of the author's analysis of the prediction of employee delay factors using three algorithms, namely the C.45 algorithm accuracy = 79.37% and AUC value = 0.646, Random Forest Algorithm accuracy = 78.58% and AUC value = 0.807 while for the Random Tree algorithm accuracy = 76.26% and the AUC value = 0.610.

RapidMiner is a software for machine learning, data mining, predictive analytics, and business analytics. The server will record large web log files when user visits the website. Extracting knowledge from such huge data demands for new methods. In this paper, we propose a web usage mining method with RapidMiner. At first, the redundant files in log file are deleted by Matlab and then we mines web log which has been pretreated with RapidMiner, it obtains the custom of different user to visit the website by processing and analyzing log file, and mines unusual rules, and provides the reference for the policy decision and construction of website. Experimental result analysis show that, applying RapidMiner in web usage mining, will obtain frequent model which user visits the website, manage to optimize the website structure and recommends for users.

By applying RapidMiner workflows has been processed a dataset originated from different data files, and containing information about the sales over three years of a large chain of retail stores. Subsequently, has
been constructed a Deep Learning model performing a predictive algorithm suitable for sales forecasting. This model is based on artificial neural network –ANN- algorithm able to learn the model starting from
sales historical data and by pre-processing the data. The best built model uses a multilayer neural network together with an “optimized operator” able to find automatically the best parameter setting of the
implemented algorithm.

This paper analyses the use of data mining techniques in evaluating job readiness of unemployed population in Ireland. To effectively help a jobseeker to enter or return to employment, it is necessary to develop a personal plan and provide them with suitable services. This report investigates how employment and further education needs are recognized among the customers of the Irish public employment services. Following the steps of CRISP-DM methodology, it explores the characteristics of the group and attempts to identify the underlying pattern. Finally, after applying suitable mining techniques, it is discussed whether the classification system with regard to job readiness can be automated.

Internet usage has become intensive during the last
few decades; this has given rise to the use of email
which is one of the fastest yet cheap modes of
communication. The growing demand of email
communication has given rise to the spam email
which is also known as unsolicited mails. In this
paper we propose an ensemble model that uses
majority voting on top of several classifiers to
detect spam. The classification algorithms used for
this purpose are Naïve Bayesian, Support Vector
Machines, Random Forest, Decision Stump and kNearest Neighbor. Majority voting generates the
final decision of the ensemble by obtaining major
votes from the classifiers. The sample dataset used
for this task is taken from UCI and the tool
Rapidminer is used for the validation of the results.

Apple adalah perusahaan teknologi multinasional yang berpusat di Cupertino, California, yang merancang, mengembangkan, dan menjual barang elektronik konsumen, perangkat lunak komputer, dan layanan daring. Sumber pendapatan Apple secara garis besar dapat dibagi menjadi lima sektor, yakni: iPhone, Mac, iPad, Aksesoris, dan Layanan (App Store, iTunes, dan lainnya). Berdasarkan data statistik, layanan merupakan sumber pendapatan terbesar kedua selama beberapa tahun ke belakang untuk Apple.

Melihat dari fakta tersebut penulis mendapatkan ide untuk melakukan operasi data mining terhadap dataset aplikasi yang terdaftar pada App Store untuk menemukan pola dan inovasi yang dapat ditawarkan terhadap perusahaan.

Dengan menggunakan teknik data mining asosiasi dengan metode matriks korelasi, penulis menemukan adanya pola dari data yang menunjukan adanya korelasi antar atribut yang berpotensi untuk meningkatkan pendapatan Apple melalui App Store.

Abstrak: Feature Extraction adalah teknik pengambilan ciri / feature dari suatu bentuk yang nantinya nilai yang didapatkan akan dianalisis untuk proses selanjutnya. Klasifikasi adalah proses untuk menyatakan suatu objek ke dalam salah satu kategori yang sudah didefinisikan sebelumnya. Dalam penelitian ini membahas tentang pengambilan ciri dari suatu feature dengan dataset yang berisi kumpulan foto wajah yang akan digunakan untuk klasifikasi pengenalan wajah pria atau wanita. Dataset yang digunakan merupakan dataset global yang diambil dari " http://vis-www.cs.umass.edu/lfw ". Dari Feature Extraction yang dilakukan diperoleh 6 atribut independent dan 1 atribut dependent. Tujuan dari penelitian ini adalah untuk membandingkan performance dari 2 algoritma yaitu Support Vector Machine dan k-Nearest Neighbor. Dari percobaan yang sudah dilakukan menunjukkan nilai rata-rata akurasi dari Support Vector Machine sebesar 58.13% sedangkan nilai rata-rata akurasi dari k-Nearest Neighbor sebesar 54.4%.

This paper introduces a three-step methodology of identifying skills demand on labour markets. By accessing publicly available vacancy data, with web and text mining tools, we are able to extract valuable facts about competences and abilities sought by employers. This easily applicable technique provides a new dimension in labour market research. It supplements occupational analysis with detailed information related to specialised knowledge expected from prospective employees. By example of IT jobs publicised through IrishJobs.ie domain, we present how web based data can be successfully acquired and pre-processed to suit our research needs, and how meaningful information can be extracted in efficient and reasonably quick manner. Evidence obtained through this process is a straightforward reflection of current needs, and can be acted on by educational and labour market bodies to bridge the gap between skills demand and supply.

According to World Health Organization (WHO), breast cancer is the top cancer in women both in the
developed and the developing world. Increased life expectancy, urbanization and adoption of western
lifestyles trigger the occurrence of breast cancer in the developing world. Most cancer events are
diagnosed in the late phases of the illness and so, early detection in order to improve breast cancer
outcome and survival is very crucial.
In this study, it is intended to contribute to the early diagnosis of breast cancer. An analysis on breast
cancer diagnoses for the patients is given. For the purpose, first of all, data about the patients whose
cancers’ have already been diagnosed is gathered and they are arranged, and then whether the other
patients are in trouble with breast cancer is tried to be predicted under cover of those data. Predictions of
the other patients are realized through seven different algorithms and the accuracies of those have been
given. The data about the patients have been taken from UCI Machine Learning Repository thanks to Dr.
William H. Wolberg from the University of Wisconsin Hospitals, Madison. During the prediction process,
RapidMiner 5.0 data mining tool is used to apply data mining with the desired algorithms.

Peningkatan kualitas karyawan yang bekerja didalam sebuah perusahaan merupakan hal penting yang harus dilakukan salah satunya dengan cara penilaian kinerja karyawan serta memberikan apresiasi terhadap karyawan yang berprestasi agar dapat memacu kinerja karyawan lainnya. Penilaian perlu dilakukan terhadap seluruh karyawan dan meliputi semua aspek sehingga menghasilkan penilaian yang obyektif, maka dari itu diperlukan otomatisasi penggalian data yang akurat dari data karyawan untuk menjadi acuhan dalam penetapan keputusan. Penerapan data mining dengan metode Algoritma C4.5 dipilih dalam penelitian ini untuk memprediksi karyawan berprestasi karena dapat mengeksplorasi data dalam jumlah yang besar dan mengubahnya menjadi pohon keputusan yang mempresentasikan aturan dengan pengembangan metode menggunakan alat bantu rapidminer. Hasil dari penelitian ini adalah diketahui akurasi sebesar 75.67% dari data karyawan. Hal ini menunjukan bahwa kinerja karyawan dalam aspek atribut-atribut berpengaruh didalam aturan yang dihasilkan pohon keputusan Algoritma C4.5 masih rendah. Oleh karena itu, diperlukan penambahan atribut-atribut berpengaruh serta penggunaan metode-metode data mining lainnya agar mendapatkan tingkat akurasi yang lebih tinggi

The internet world expands day by day as well as threats related to it. Nowadays, Cyber-attacks often happen more than a decade ago. Intrusion detection is one of the most popular search area that provides various technologies and security techniques for detecting cyber-attacks. Different data extraction tools learn to algorithms that help in the implementation of the Learn to build Identities. In this paper, we have done a comparative study for machine learning tools using WEKA and Rapid Miner with two algorithms Random Tree and Random Forest for network intrusion detection. These can be used to implement intrusion detection techniques based on data mining. Analysis of the initial results of two different machine learning tools WEKA and Rapid Miner is carried out using KDD' 99 attack dataset and results are the best tools is WEKA, while the best algorithms is Random Forest.

Abstract. The experiments aimed to compare machine learning algorithms to create models for the valuation of residential premises, implemented in popular data mining systems KEEL, RapidMiner and WEKA, were carried out. Six common methods comprising two neural network algorithms, two decision trees for regression, and linear regression and support vector machine were applied to actual data sets derived from the cadastral system and the registry of real estate transactions. A dozen of commonly used performance measures was applied to evaluate models built by respective algorithms. Some differences between models were observed.

Accounting databases with fraudulent transactions inside was used to detect fraud patterns by data mining tool. The object was accomplished by the following method: first, inside data, fraudulent transactions according to three fraud patterns were settled, over it the algorithms, Euclidian distance and local outlier factors were run using Rapidminer program. As a result the fraud patterns were shown in different ways according to the specific graphics contributed by the program. In conclusion, clusters grouping by Euclidian distance with k Means algorithm (k = 4) allowed an adequate visualization of the values’ distribution, as consequence was detected the first and third fraud patterns. The application of the outlier’s detection algorithm (LOF) detected the three fraud patterns in a clear way as a consequence of the insolate outliers in different graphics, shown by Rapidminer program, with different variables correlated.

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.

Using the descriptive and developmental design, the study developed and evaluated an eHealth application for women using decision tree classifiers. It focuses on the development of an eHealth application system using open-access datasets from UCI Machine Learning Repository. This attempts to predict the onset of diabetes and chronic kidney diseases grounding from the generated predictive models. These decision models are created using C4.5, ID3 and CART algorithms with RapidMiner data science platform. Performance metrics are deployed such as accuracy, recall, precision and error rate to compare the reliability of each model. Models incurred the highest assessment are the bases of the developed system following Agile Software Development Life Cycle Model. Easy access to healthcare workers through teleconsultation, diabetes and chronic kidney disease (CKD) online diagnosis, and maternal care videos are possible with this study. The summary of the evaluation showed that the eHealth Application got an overall average weighted mean of 3.98, which is described as high extent. Based on the respondents' response, the strongest point of the system was its portability, which earned the highest average mean among categories of system evaluation. Thus, the system addresses the shortcomings of healthcare in terms of distance and timeliness of treatment fostering an equal access to healthcare.

Research on the performance of workers based on the determination of standard working hours for absences conducted by workers in a certain period. In disciplinary supervision, workers are expected to be able to provide the best performance in the implementation of work in accordance with predetermined working hours. The measurement of the level of discipline of admission hours for placement workers is carried out every working day, continuously and continuously. Attendance monitoring already uses online attendance by using data downloaded from the online attendance provider as the main data. In addition, data collection is done by filtering employee absentee data and supporting information on the categories that cause mismatches in meeting work schedules. Mobilization of workers according to location and working hours has been regulated in company regulations allowing the placement of workers in accordance with the residence so as not to affect the desired work results the company is still within reasonable limits and can be increased. The assessment of this study as a progression factor inhibiting the company in achieving company targets. From the results of the author's analysis of the prediction of employee delay factors using three algorithms, namely the C.45 algorithm accuracy = 79.37% and AUC value = 0.646, Random Forest Algorithm accuracy = 78.58% and AUC value = 0.807 while for the Random Tree algorithm accuracy = 76.26% and the AUC value = 0.610.

According to World Health Organization (WHO), breast cancer is the top cancer in women both in the developed and the developing world. Increased life expectancy, urbanization and adoption of western lifestyles trigger the occurrence of breast cancer in the developing world. Most cancer events are diagnosed in the late phases of the illness and so, early detection in order to improve breast cancer outcome and survival is very crucial. In this study, it is intended to contribute to the early diagnosis of breast cancer. An analysis on breast cancer diagnoses for the patients is given. For the purpose, first of all, data about the patients whose cancers' have already been diagnosed is gathered and they are arranged, and then whether the other patients are in trouble with breast cancer is tried to be predicted under cover of those data. Predictions of the other patients are realized through seven different algorithms and the accuracies of those have been given. The data about the patients have been taken from UCI Machine Learning Repository thanks to Dr. William H. Wolberg from the University of Wisconsin Hospitals, Madison. During the prediction process, RapidMiner 5.0 data mining tool is used to apply data mining with the desired algorithms.

In this paper the authors proposed different Multilayer Perceptron Models (MLP) of artificial neural networks (ANN) suitable for visual merchandising in Global Distribution (GDO) applications involving supermarket product facing. The models are related to the prediction of different attributes concerning mainly shelf product allocation applying times series forecasting approach. The study highlights the range validity of the sales prediction by analysing different products allocated on a testing shelf. The paper shows the correct procedures able to analyse most guaranteed results, by describing how test and train datasets can be processed. The prediction results are useful in order to design monthly a planogram by taking into account the shelf allocations, the general sales trend, and the promotion activities. The preliminary correlation analysis provided an innovative key reading of the predicted outputs. The testing has been performed by Weka and RapidMiner tools able to predict by MLP ANN each attribute of the experimental dataset. Finally it is formulated an innovative hybrid model which combines Weka prediction outputs as input of the MLP ANN RapidMiner algorithm. This implementation allows to use an artificial testing dataset useful when experimental datasets are composed by few data, thus accelerating the self-learning process of the model. The proposed study is developed within a framework of an industry project.

Floor coverings are required to be non-slip so that people can move freely in public and closed areas. Within the scope of this study, dry and wet environment Dynamic Friction Coefficients of floor coverings DCOF (µ) are measured in 5 different areas (Entrance, Corridor, Room, Staircase, Sink) in 7 different public institutions by using GMG 200 test device according to DIN 51131:2014-03, EN 13893 and TS CEN /TS 16165 Annex-D standards. The dynamic friction coefficient DCOF (µ) values obtained with the arithmetic mean were grouped by using the K-Means Method in the RapidMINER program, and the boundary values of the cluster groups formed were determined by using the Decision Tree Method. The results were evaluated using the Wuppertaler Signage which is currently used in the German Accident Insurance System and is in the current DIN 51131:2014-03, EN 13893 and TS CEN/TS 16165 Annex-D Standards. As a result of the evaluation, an anti-slip risk map of public institutions was drawn and the safest institution was determined as K5 SGK institution, while the most unsafe institution was determined as K3 Kindergarten. It has been determined that the friction coefficients of ceramic floor coverings, which are used as floor coverings in 7 different public institutions, are low and it is recommended to replace these floor coverings with materials with high dynamic friction coefficient or to take necessary preventive and protective measures.
Topluma açık ve kapalı alanlarda insanların rahat hareket edebilmeleri için zemin kaplamalarının kaymaz özellikte olması istenmektedir. Bu çalışma kapsamında 7 farklı kamu kurumunda 5 farklı alanda (Giriş, Koridor, Oda, Merdiven, Lavabo) zemin kaplamalarının kuru ve ıslak ortam dinamik sürtünme katsayıları DCOF (μ) GMG 200 test cihazı kullanılarak DIN 51131:2014- 03, EN 13893 ile TS CEN/TS 16165 Ek- D standartlarına göre ölçülmüştür. Aritmetik ortalama ile elde edilen dinamik sürtünme katsayısı DCOF (μ) değerleri RapidMİNER programında K-Ortalamalar Yöntemi kullanılarak gruplanmış, oluşan küme grupları sınır değerleri Karar Ağacı Yöntemi kullanılarak belirlenmiştir. Sonuçlar günümüzde Alman Kaza Sigortası Sisteminde kullanılan ve güncel DIN 51131:2014- 03, EN 13893 ve TS CEN/TS 16165 Ek- D Standartlarında geçen Wuppertaler Tabelası kullanılarak değerlendirilmiştir. Değerlendirme sonucu kamu kurumlarına ait kaymazlık risk haritası çıkarılmış olup, en güvenli kurum K5 SGK kurumu olurken en güvensiz kurum ise K3 Ana okulu olarak belirlenmiştir. 7 farklı kamu kurumunda da zemin kaplaması olarak kullanılan seramik zemin kaplamalarının sürtünme katsayılarının düşük olduğu tespit edilmiş ve bu zemin kaplamalarının dinamik sürtünme katsayısı yüksek malzemelerle değiştirilmesi veya gerekli önleyici ve koruyucu tedbirlerin alınması önerilmiştir.

Accounting databases with fraudulent transactions inside was used to detect fraud patterns by data mining tool. The object was accomplished by the following method: first, inside data, fraudulent transactions according to three fraud patterns were settled, over it the algorithms, Euclidian distance and local outlier factors were run using Rapidminer program. As a result the fraud patterns were shown in different ways according to the specific graphics contributed by the program. In conclusion, clusters grouping by Euclidian distance with k Means algorithm (k = 4) allowed an adequate visualization of the values’ distribution, as consequence was detected the first and third fraud patterns. The application of the outlier’s detection algorithm (LOF) detected the three fraud patterns in a clear way as a consequence of the insolate outliers in different graphics, shown by Rapidminer program, with different variables correlated.

Abstract.
This paper presents a study about three software tools for data mining which are commonly utilized as for commercial as for scientific purposes. Due to the existence of a variety of tools, many novice researchers in the subfield of data mining have problems to choose which tool is more appropriate to be used in their research. Thus, the purpose of this paper is to present a brief survey about the following tools: KEEL, RapidMiner, and WEKA.

Resumo:
Este artigo apresenta um estudo sobre três ferramentas de software utilizadas para mineração de dados as quais são referencias tanto para uso comercial quanto para uso científico. Devido a existência de uma variedade de ferramentas, muitos pesquisadores iniciantes em mineração de dados têm dificuldade para escolher a ferramenta adequada a ser utilizada em suas pesquisas. Portanto, o objetivo deste trabalho é apresentar uma breve pesquisa sobre as seguintes ferramentas: KEEL, RapidMiner e WEKA.

By applying RapidMiner workflows has been processed a dataset originated from different data files, and containing information about the sales over three years of a large chain of retail stores. Subsequently, has been constructed a Deep Learning model performing a predictive algorithm suitable for sales forecasting. This model is based on artificial neural network-ANN-algorithm able to learn the model starting from sales historical data and by pre-processing the data. The best built model uses a multilayer neural network together with an "optimized operator" able to find automatically the best parameter setting of the implemented algorithm. In order to prove the best performing predictive model, other machine learning algorithms have been tested. The performance comparison has been performed between Support Vector Machine-SVM-, k-Nearest Neighbor k-NN-,Gradient Boosted Trees, Decision Trees, and Deep Learning algorithms. The comparison of the degree of correlation between real and predicted values, the average absolute error and the relative average error proved that ANN exhibited the best performance. The Gradient Boosted Trees approach represents an alternative approach having the second best performance. The case of study has been developed within the framework of an industry project oriented on the integration of high performance data mining models able to predict sales using-ERP-and customer relationship management-CRM-tools.

Internet usage has become intensive during the last few decades; this has given rise to the use of email which is one of the fastest yet cheap modes of communication. The growing demand of email communication has given rise to the spam email which is also known as unsolicited mails. In this paper we propose an ensemble model that uses majority voting on top of several classifiers to detect spam. The classification algorithms used for this purpose are Naïve Bayesian, Support Vector Machines, Random Forest, Decision Stump and k-Nearest Neighbor. Majority voting generates the final decision of the ensemble by obtaining major votes from the classifiers. The sample dataset used for this task is taken from UCI and the tool Rapidminer is used for the validation of the results.

Smart Watches, mobile computers in the format of a wristwatch and worn on the wrist, are generally estimated as the next mega trend of mobile computing. Meanwhile, »wearables« are available in the mass market1, IP67 certified, which feature an autonomous operating 3G mobile radio and include AAL relevant sensors like microphone, 3D-accelerometer and gyroscope, magnetometer, barometer, GPS and heart rate monitor. In addition to the 3G mobile radio, the chosen smart watch can communicate via Bluetooth 4.0 LE or WLAN. In this paper, first we describe the sensoric and commu-nicative potential of smart watches and then present the structure and practical experiences of and with our smart watch based solution for AAL. Subsequently, we compare the requirements in this sector with the current offering of commer-cially available products.