International Journal of Engineering and Emerging Technology, Vol. 2, No. 1, January—June 2017 (p-issn: 2579-5988, e-issn: 2579-597X) 62 Bussines Intelligent in Telemarketing Using SVM Putu Agung Ananta Wijaya [1] , Komang Budiarta [2] , and Made Sudarma[3] [1][2] Department of Electrical and Computer Engineering, Post Graduate Program, Udayana University [3] Department of Electrical and Computer Engineering, Udayana University Email: ananta_wijaya@student.unud.ac.id Abstract— Direct marketing provides an advantage in approaching consumers. Communication that happens allows us more closely, able to change the behavior and know the needs required by consumers accurately. But this technique has a lack of time. It takes a long time to convince consumers to buy the products offered. Bussines intelligent with data mining approach to consumer data is required. This process will analyze the potential possessed by a consumer. At the stage of the DSS used SVM method to predict whether consumers will buy products that have been offered. Bussines intelligent built proven able to predict consumers who have the potential to buy products. Tests show the greatest prediction accuracy rate is 89.5% with a combination of data traning of 70% of the dataset. Index Term—Data Mining, Linier SVM, Holdout I. INTRODUCTION Marketing is a series of activities to plan, promote, and distribute goods or services to consumers in order to meet the needs of consumers.Marketing is about identifying and meeting human and social needs[1]. One of the shortest good definition of marketing is meeting needs profitiably. To convey the value of the product to the customer is required communication. Marketing communications are the means by which the company in an effort to inform, persuade, and remind consumers directly or indirectly about the product or brand they sell [1]. Communication method on marketing is divided into two, indirect marketing and direct marketing. Indirect marketing is a strategy to promote a product or service intended to touch the mind and feelings of consumers indirectly. Indirect sales can be found in the form of advertising, corporate social responsibility, and interactive marketing via the internet. Direct Marketing is an interactive marketing system that utilizes one or more advertising media to generate measurable responses and or transactions in any location [2]. One form of direct marketing is telemarketing [3]. Direct marketing provides an advantage in approaching consumers. Communication that happens allows us more closely, able to change the behavior and know the needs required by consumers accurately. With direct marketing the company will more quickly know the results of marketing done. But this technique has a lack of time. It takes a long time to convince consumers to buy the products offered. Business intelligence with data mining approach to consumer data is needed. Business Intelligence is comprises the set of strategies, processes, applications, data, technologies and technical architectures which are used by enterprises to support the collection, data analysis, presentation and dissemination of business information[4]. Data mining is the analysis of (often large) observational data sets to find unsuspected relationships and to summarize the data in novel ways that are both understandable and useful to the data owner[5]. The system to be built has a purpose to analyze the potential possessed by a consumer. In the DSS stage used SVM method (Support Vector Machines) to predict whether consumers will buy the products that have been offered. SVM is a pattern recognition method that is classified as supervised. In a two-class learning task, the aim of SVM is to fi nd the best classi fi cation function to distinguish between members of the two classes in the training data [6]. This method offers great resilience and accuracy among other popular methods [7]. With the construction of this system is expected to make a decision about whether or not the marketing process can be taken faster. And when it comes to consumers who have the potential to buy products then it can be compiled more quickly the appropriate marketing strategy. II. RELATED RESEARCH A. Prediksi Keputusan Klien Telemarketing Untuk Deposito Pada Bank Menggunakan Algoritma Naive Bayes Berbasis Backward Elimination In this research will analyze telemarketing dataset from UCI Repository. The method of classification used is the Naive Bayes algorithm. This method has a purpose to predict the decision of Telemarketing clients. The method of classification will be combined with the feature selection method. The feature selection method used is Backward Elimination. The test results show that the accuracy of Naive Bayes 89.08%, after the selection of features using Backward Elimination obtained a higher accuracy of 90.69%, by looking at the accuracy of the algorithm Naive Bayes-based Backward Elimination increase accuracy to predict Decision of Telemarketing clients [8]. B. Optimasi Algoritma Support Vector Machine (SVM) Menggunakan Adaboost Untuk Penilaian Risiko Kredit Support Vector Machine (SVM) is an algorithm proposed by many researchers in the field of data mining credit risk. But the difficulty of determining the parameters of an ideal is the problem of research in improving the accuracy of SVM. In this