A hybrid cost estimation framework based on feature-oriented data mining approach

A data mining based load forecasting strategy for smart electrical grids

“…Filter approach is computationally simple and fast. Due to its independence of the mining algorithm, the features selection needs to be performed only once, and then different classifiers can be assessed [7]. Examples of the filter based approaches are given in Table 1.…”

“…Many data mining algorithms don't perform well with those irrelevant attributes or features. Therefore, features selection techniques are used to remove those non-informative features before the use of the mining algorithm [7]. The main objectives of features selection techniques are to avoid overfitting and improve the performance of the model, which leads to faster and more cost-effective models [7].…”

“…Several prediction algorithms don't perform well with large amounts of features, while many ineffective features may be presented in electrical load data [7]. Feature selection has a good impact in improving the model performance as well as providing faster decisions as it minimizes the considered features to the only effective ones [7]. On the other hand, training of the forecasting model with raw data that usually contains outliers certainly degrades the system's performance as the decisions are affected by those rare data whose behavior is very exceptional.…”

“…[6]. Several prediction algorithms don't perform well with large amounts of features, while many ineffective features may be presented in electrical load data [7]. Feature selection has a good impact in improving the model performance as well as providing faster decisions as it minimizes the considered features to the only effective ones [7].…”

Integrating Data Mining Into Managerial Accounting System: Challenges and Opportunities

“…The final proposed model could be repeatable, robust, and accurate. Besides, data mining needs to be supported with available and comprehensive business or manufacture process data for implementation (Sajadfar & Ma, 2015).…”