pmanlukas/thesisdata

Thesis Data 0.0.1

(c) by Lukas Pollmann

This Repository is part of my master thesis

In the local version the other repos are also part of the disk

Thesis data is a data set created for my master thesis on the classification of openAPI specification JSON Files.

You will find all relevant files of my design research process

⛔ The code for the PoC of the classifier can be found in this repository

⛔ The code for the linear SVM based classifier for benchmarking can be found in this repository

⛔ The code for an example implementation of the classifier as api can be found in this repository

⛔ The code for the static web page for the dataset can be found in this repository

Here a short discription on the Notebooks:

• Analyse Cluster (Diff): notebooks that analyze the results of the clustering
• classes verification: notebook showing the underlying samples of the clusters resulting from the clustering
• Data exploration - paths: notebook analyzing the path objects of each sample and creating dataframes for clustering
• clustering: notebook clustering the data set samples regarding structure and its content
• combine dataframes: notebook showing results of the clustering in R and comparing the results of each approach
• Data exploration - completness: notebook analyzing the completeness of fields of the samples
• cortical api tests: notebook playing around with the cortical API
• dataset labels creation and export: notebook consolidating the results of clustering in classes and uploading the resulting data sets to blobs
• category playground: notebook to examine the categories of the dataset
• versions: notebook to check and compare versions of the samples in the dataset
• variance_cluster: notebook to check the variance between the different versions of the data set samples and differences in the clusters
• download specs: notebook used to download the raw data for the dataset
• get_data_apiguru: notebook used to explore and interact with the api guru database api

Here a short discription on the folders:

• .idea: folder of Pycharm (only binaries, will be deleted)
• R - scripts clustering: folder will contain all R scripts and files used in project
• clustering results: folder containing all files and results of clustering approaches
• notebooks html: the html version of the used Jupyter Notebooks
• obj: folder containing all pickle objects used in the design process
• specs: folder containing all data regarding the specifications
• visualization: folder containing other visualizations
• enviroment: folder containing the used anaconda enviroment as yaml file

The other files are currently still in active usage. Over time they are consolidated in existing or new folders!