I am trying to assess or predict the use of one specific technology that might have several impacts on one operational activity. This technology is supposed to improve the efficiency of the activity in various aspects among which the economic aspects are of high importance to me. I am trying to use a technique in Big Data Analytics that can help me measure and predict the economic outcomes of this change. First, I am going to analyze the historic big data and find KPIs. Second, I would use a technique like deep learning to predict the effects of the change made by using the new technology and check how it can change the KPIs.
But now what I need is to know how I can combine this technique with another quantitative technique in order to ensure the accuracy and precision of the results. I would like to make and use a hybrid model. It might be a technique in fields such as:
- forecasting,
- technology impact assessment,
- data science,
- statistics,
- economic impact analysis,
- decision science,
- econometrics,
- techno-economic analysis,
or any related field.
I would be also thankful if you could share with me any research articles helping me know more about this type of research methodology.
You can use algorithm in each sector of following
feature selection/feature engineering +
classifier for model+
tune model for Increase performance/ prune the network for simplification.
Have a look at https://www.getsmarter.com/blog/career-advice/big-data-analysis-techniques/ and https://www.firmex.com/resources/blog/7-big-data-techniques-that-create-business-value/
Dear Masoud, I guess you can use machine learning as well as evolutionary algorithms for both predicting KPIs and optimizing. Needless to say that using deep learning can overfit the network hence you should use techniques like CV10.
You can use algorithm in each sector of following
feature selection/feature engineering +
classifier for model+
tune model for Increase performance/ prune the network for simplification.
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics