[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Subscribe]
DM: RE: Neural Network SoftwareFrom: Nina Fitzsimons Date: Mon, 31 Jul 2000 09:56:22 +0100 -----Original Message----- From: Cristina Davino [mailto:davino@dms.unina.it] Sent: 11 July 2000 15:33 To: datamine-l@nautilus-sys.com Subject: DM: Neural Network Software Dear friends, we are developing a study on neural networks software whose purpose is the improvement of the quality of neural networks software. If you are the developer or the maintainer of a Neural Networks software, we would be very pleased if you could fill the following questionnaire so as to include your software in our study and if you could supply a list (name and e-mail) of three users who might be willing to give their assessment of your software's capabilities. Please, reply by e-mail to davino@dms.unina.it or by fax to +39 081 675009 or by ordinary mail to Cristina Davino - Department of Mathematics and Statistics, University of Naples Federico II, Via Cinthia Monte S. Angelo, 80126 Naples, Italy. Thanks for cooperating. Best regards, Cristina Davino ********************************* Department of Mathematics and Statistics University of Naples Federico II Via Cinthia Monte S. Angelo 80126 Naples ITALY Tel. +39 081 675183 Fax +39 081 675009 e-mail davino@dms.unina.it ********************************* 1. Software Name __Neuframe__________________ 2. Country________UK_____________ 3. Year of the first version of the product _______ 4. Year of the latest version of the product___2000_____ 5. On which platform can the software work? |_| Dos |x_| Windows |_| Macintosh |_| Unix |_| Other _____________ 6. How can be classified the software? |_| Freeware |_| Shareware |_x| Commercial 7. What are the minimal configuration requirements? Recommended Recommended Configuration - Windows 9X, NT4 or 2000, Pentium 200 with 64Mb Ram Space ____________ RAM _______________ Processor____________ Other ________________ 8. What is the programming language used for the development of the software? ___C++_________________ 9. Has the software a Graphical User Interface? |x_|Yes |_|No 10. Does the software provide a programming language that allows to introduce user defined functions? |_| Yes if Yes, which one: ____________) |_| No The product can generate code extract of the model in C, C++ and Java 11. How can you define the actual state of the product? |_x| finished |_| under development and innovation |_| under total changements 12. Is the software part of some general purpose software? |_| Yes |x_| No 13. Which is the maximum data sets size? Number of rows: Number of columns: This is dependent on the type data, numerical, categorical etc, number of variables, algorithm chosen etc. However we tested our software on data of 10,000 rows and 300 columns. 14. Which kind of data can be imported? |x_| Text/ASCII |x_| Excel |_x| Access |x_| Other __ODBC _________ 15. Does the software provide the following utils? |x_| a data sheet that allows to input the data |_| export of tables |_| export of graphs |_| print of output tables |_| print of output graph 16. What are the main distribution and support features of the software? |_x| on-line help |x_| users'manual |x_| technical support available |x_| demo and examples |_| guided installation program 17. Is it present an intelligent checking of user actions in your software? |_| Yes |_x| No 18. Which kind of learning algorithms are implemented? |_| Supervised |_| Unsupervised |x_| Supervised and Unsupervised 19. How many learning algorithms are implemented? __4_________ 20. Does your software provide Pre-processing techniques? |x_| Yes (if Yes, Which ones: ___Cellcheck (checking for outliers et, data partitioning and encoding_________________________) |_| No 21. Does the software include missing values treatment procedures? |x_| Yes (if Yes, Which ones: ___as above_________________________) |_| No 22. Does your software provide automatic procedures for the selection of the architecture? |_| Pruning Which ones _______________ |_| Decision Trees |_| Others _________________ 23. Which kind of transfer functions are implemented? |_x| linear |_| logistic |_x| tangent |_x| hyperbolic |_x| Other sygmoid, gaussian, cauchy,inverse multi quadric________ 24. Which kind of training styles are implemented? |_x| Incremental training |_x| Batch training 25. Which kind of net input functions are implemented? |_| product |_| sum |_| Other ___________ 26. Does the software provide improving generalization techniques? |_|Regularization |x_| Early Stopping |_| Other ___________ 27. Which kind of performance measures are implemented? |_| Mean Square Error |_| Mean Absolute Error |_x| Root Mean Square Error |_| Other __________________ 28. Does the software allows to introduce user defined: |_| transfer functions |_| net input functions |_| performance measures 29. Does the software provide validation of the results techniques? |_x| Cross-validation (rbf) |_| Bootstrap |_| Jackknife |_| Others ______ 30. What are the main extendibility features of the software? |_| possibility to add user defined learning algorithms |_| possibility to add user defined transfer functions |_| possibility to eliminate some weights of the net 31. Do you think the software is easy to learn? |x_| very easy |_| quite easy |_| not easy 32. Is the software configurable by the user?|_| No |_| Yes 33. Who are the software tipical users? |_x| very expert Neural Networks users |x_| quite expert Neural Networks users |x_| not expert Neural Networks users 34. What is the fields of application nearest to the software features? |_| Statistics |_| Phisics |_x| Engineering |x_| Biology |x_| Medicine |_x| Other finance, retail, marketing, manufacturing etc_______ 35. Do you think the software is suitable for: |x_| teaching |x_| research |x_| Other
|
MHonArc
2.2.0