The following information was submitted:
Transactions: WSEAS TRANSACTIONS ON COMPUTERS
Transactions ID Number: 42-273
Full Name: Vicente García
Position: Ph.D. Candidate
Age: ON
Sex: Male
Address: Av. de Vicent Sos Baynat, s/n, 12071, Castelló de la Plana
Country: SPAIN
Tel: +34 964 72 83 59
Tel prefix: +34 964 72 83 48
Fax: +34 964 72 83 59
E-mail address: vgarciaj@hotmail.com
Other E-mails: jimenezv@lsi.uji.es
Title of the Paper: Surrounding–SMOTE Algorithms for Learning from Imbalanced Data Sets
Authors as they appear in the Paper: Vicente García,J. S. Sánchez, R. A. Mollineda
Email addresses of all the authors: vgarciaj@hotmail.com,jimenezv@lsi.uji.es
Number of paper pages: 10
Abstract: Many traditional approaches to pattern classification assume that the problem classes share similar prior probabilities. However, in many real-world problems, this assumption is grossly violated. Often, the ratios of prior probabilities between classes are extremely skewed. This situation is known as the imbalance problem. One of the methods to tackle this problem consists to balancing the class by resampling the original data set. The SMOTE algorithm is a most popular method to increase the size of the minority class by adding new synthetic minority instances. From the original SMOTE algorithm, we here propose the use of three surrounding neighborhood approaches with the aim of generating artificial minority examples, but taking into account the proximity and the spatial distribution of the examples. This paper show the convenience to use surrounding methods in SMOTE over ten databases and four classifiers.
Keywords: Imbalance, Over-sampling, Surrounding neighborhood, SMOTE, Proximity graph, k-NCN
EXTENSION of the file: .pdf
Special (Invited) Session:
Organizer of the Session:
How Did you learn about congress:
IP ADDRESS: 150.128.126.126
