=================================
Dataset: Library
Domain: Library Management System
=================================

-------
Version
-------

	Version 1.0 (Dec 2017)

-----------
Description
-----------

	The library software product line (SPL) consists of a management system equipped with all the operations and facilities needed to provide services to its membership holders. In general, a library has its own management system, operating environment, payment methods, network, and security system. In addition, the library offers various services to its users through an offline and online environment. Thus, this system consists of multiple software management modules (i.e., features) that are: policy, item, customer, service, operating environment, security, and payment. 

	The product configuration is performed as a process of customization and generation of a specific system for each customer depending on their context. The configuration activity defines the functionalities or properties that a system must have that meet the customer needs to manage the library resources. The complete system is considered a large application as it is composed of 135 features and billion of possible different valid product variants.


---------------
Data statistics
---------------

    135 features
	74 real configurations
	> 10^9 possible valid configurations (S.P.L.O.T. tool <http://www.splot-research.org/>)
    	none cross-tree constraints
	5 context dimensions
		- Resource Access [Digital, Phisical]
		- Device [Mobile, Computer, Tablet, None]
		- Internet Connection [Yes, No]
		- Environment [City, Company, University, Rural, Prison, Any]
		- Age Range [<10, 10-20, 20-60, >60]

---------------------
Files and data format
---------------------
	
	The data is formatted as follows:

			
	* Library.xml
	
		This file contains the specification of the Library feature model and it can be imported in FeatureIDE tool <https://featureide.github.io/>.


	* configurations ".config"
   
		Each file contains the names of the selected features by each customer.
		The set of all files is used as our historical dataset of configurations to make predictions.
		The files can be imported in FeatureIDE tool <https://featureide.github.io/>.
		
------- 
License
-------

	The customer names and other personal information are not provided for this dataset.
	The data contained in Library.zip is made available for non-commercial use.

----------------
Acknowledgements
----------------

	This work was supported by the National Counsel of Technological and Scientific Development - CNPq (grant 202368/2014-9).

----------
References
----------
   
	<The dataset is released in the 22nd International Systems and Software Product Line Conference (SPLC'18): http://splc2018.net/>

    	When using this dataset you should cite:
	
		<@inproceedings{PSF+:SPLC18,
			title={N-dimensional Tensor Factorization for Self-Configuration of Software Product Lines at Runtime},
			author={Pereira, Juliana Alves and Schulze, Sandro and Figueiredo, Eduardo and Saake, Gunter},
			booktitle={Proceedings of the International Systems and Software Product Line Conference(SPLC)},
			pages={?},
			year={2018},
			organization={ACM}
		}>

-------
Credits
-------

	The dataset of configurations was built by 37 Software Engineering and Database Master and PhD students from two universities: 
		- University of Magdeburg, Germany 
		- University of Minas Gerais, Brazil.
   
-------   
Contact
-------

   	Juliana Alves Pereira, juliana.alves-pereira@ovgu.de