Multiclass IFROWNN Classification Algorithm Using OVA and OVO Strategy
Shital Andhalkar, B. F. Momin · 2018
All basic algorithms are commonly used for decision making, in the field of machine learning. But basic algorithms are used to handle more often balanced data. Now a day, enormous people using Artificial intelligence in their life. For example, IOT devices uses for home automation, security etc. Because of that imbalanced data generates rapidly. But handling this data is big challenge and specifically handling multiclass imbalance data is more challenging. In this paper, we deal with multiclass datasets and to achieve more correct result we convert multiclass problem to multiple binary class problem using OVA (one vs all) and OVO (one vs one) methodology.