Forecasting and data mining can become particularly challenging when data:

• Is particularly large or complex
• Has limited history
• Contains a large amount of irrelevant data
• Has pockets of missing information

Advanced solutions may be more cost-effective for cases where data is particularly challenging. Examples of advanced techniques include:

• Auto_ARIMA: Useful for situations where the data is prone
  to seasonality “spikes” or “outliers”.
• Neural Networks: A technique that takes advantage of repetitive
  learning through repeated forecasts and comparisons
  against assumptions. This technique is suited to situations
  where data contains missing values, short time periods,
  inconsistent reporting of data and where the user has multiple
  items to predict
• Genetic Algorithm: Popular for solving optimization, search and machine learning problems.
• Naïve Bayes: A Naïve Bayes classifier can be trained to classify patterns involving thousands of attributes and applied to thousands of patterns. As a result, Naïve Bayes is a preferred algorithm for text mining and other large classification problems.