Using Artificial Neural Networks to Provide Guidance in Extending PL/SQL Programs

This project involves the creation of an artificial neural network model to automatically predict the correct placement of a new PL/SQL object among one of the existing schemas. The model uses dependencies among the software and database objects as features for training. Then, given a new object and the list of other objects it uses, the network can predict the schema, where the object should be included.

This repository includes executable Python scripts in the form of Jupiter notebooks for creating the model, optimizing the hyperparameters of the model via grid search, training and validating the model. It also includes a set of datasets used for training and validating the model as well as the obtained results.

Grid search for optimizing the hyperparameters is performed by using the Scikit-Learn-GridSearchCV module. The corresponding code for one of the optimized parameter sets (as listed below) is provided in CRM_ALL_Part2.ipynb and CMS_ALL_Part2.ipynb.

  'hidden_layer_sizes': 
    (50, 50),
    (50, 100),
    (50, 150),
    (50, 200),
    (50, 250),
    (50, 500),
    (50, 750),
    (50, 1000),
    (50, 1250),
    (50, 1500),
    (50, 1750),
    (50, 2000),
    (50, 2250),
    (50, 2500),
    (50, 2750),
    (50, 3000)
   'activation': 'logistic', 'tanh'
   'alpha': 0.0001, 0.0005, 0.001, 0.0015, 0.002

The following two files can be used for reproducing our study and results.

• CRM_ALL_Reproduce_Notebook.ipynb is used for performing prediction based on the CRM-ALL dataset (CRM_ALL_Dataset_x.p).
• CMS_ALL_Reproduce_Notebook.ipynb is used for performing prediction based on the CMS-ALL dataset (CMS_ALL_Dataset_x.p).

Top_10_Results_with_Splits.xlsx includes top 10 results for all datasets with split scores.

All datasets (14 files) are shared as pickle dump with the filenames listed as follows.

• CMS_ALL_Dataset_x.p
• CMS_ALL_Dataset_y.p
• CMS_T_Dataset_x.p
• CMS_T_Dataset_y.p
• CMS_TVPF_Dataset_x.p
• CMS_TVPF_Dataset_y.p
• CRM_ALL_Dataset_x.p
• CRM_ALL_Dataset_y.p
• CRM_T_Dataset_x.p
• CRM_T_Dataset_y.p
• CRM_TV_Dataset_x.p
• CRM_TV_Dataset_y.p
• CRM_TVPF_Dataset_x.p
• CRM_TVPF_Dataset_y.p

Environment and Setup

• *The Python version is 3.6.3.

• Employed Python Libraries:

  - *Pickle version is 4.0.
  - *Numpy version is 1.13.3.
  - *Scikit-learn version is 0.19.1.
  - *Pandas version is 0.20.3.
  

How To Run

1. Install anaconda (It will install all the necessary tools and libraries for you) https://www.anaconda.com/products/individual
2. Download all the files on github to a local folder
3. Run "Anaconda Prompt (Anadonda3)"
4. Go to the local folder where the project files are downloaded
5. Type "jupyter notebook" in the command line