[ Home | Submit ]
[ Humor | Icons | Library | Manuals | Organizations | RFCs | Services | Software | Users ]
| Version: | 3.1 |
| Summary: | Neural net & DLL development kit for Win 3.1 |
| License: | Shareware |
| Requires: | Microsoft Word (or Word Viewer). |
| Email: | ainet@siol.net |
| Homepage: | http://www.ainet-sp.si/ |
| Download: |
ain16125.zip (Sep 27 1997, 1.7M) |
| Description: |
A neural network-like application. It does not require training samples, has no limits in number of problem variables (input & output neurones), no limits in sample size base. It is not sensitive to noise in the data. It provides several different charts types and comes with a manual written in Microsoft Word. It is very easy to use, due to the build-in spreadsheet-like editor. It is fast and efficient and comes with several examples, which are solved and explained in the manual. It has a fully documented DLL library which can be used to create your own applications on the basis of aiNet algorithm. Any language which can make calls to DLL libraries can be used. |
| Summary: | Backpropagation and quickprop neural network |
| License: | Freeware for individuals, Shareware for businesses and government.
|
| Download: |
backpro1.zip (May 10 1999, 154.9K) |
| Description: |
A neural networking program that implements backpropagation and quickprop, suitable for certain types of pattern recognition and prediction tasks. The software was originally done for my introduction to Artificial Intelligence textbook, The Pattern Recognition Basis of Artificial Intelligence. |
| Summary: | Instance-based learning system |
| License: | Shareware |
| Author: | Steve G. Romaniuk (Steve G. Romaniuk) |
| Email: | todd@alberts.com |
| Homepage: | http://www.alberts.com/ |
| Download: |
mpil10.zip (May 28 1997, 134.8K) |
| Description: |
MPIL is an instance-based learning system, which utilizes two models for creating neighborhoods. The first model places a single neighborhood sphere (based on Euclidean distance measure) around an instance, and is in nature similar to the nearest neighbor classifier, except that it removes redundant instances. The second model incorporates N radii (one for each input of an instance). This model also supports knowledge acquisition in the form of rule extraction. In a sense, both approaches are similar to neural networks in that they exploit a very similar parallelism. MPIL represents a good alternative in cases were large amounts of data have to be learned and provides good facilities for storage reduction. |
| Summary: | Backprop network for Win3.1 with Pascal source |
| Download: |
neural10.zip (Jun 7 1993, 166.9K) |
| Summary: | Create, Edit, Train & Test Neural Networks |
| License: | Shareware |
| Author: | Stephen Wolstenholme, Neural Planner |
| Email: | steve@tropheus.demon.co.uk |
| Homepage: | http://www.tropheus.demon.co.uk/ |
| Download: |
np452s.zip (Jun 16 1998, 628.7K) |
| Description: |
Neural Planner is a neural network system for Microsoft Windows. It is for people who want to use neural networks. It is not aimed at people who want to investigate or research into neural networks, it is not a tool kit. It allows the user to produce multilayered neural networks using a simple graphical editor or create standard three layer networks from training files. The user can also produce training, testing and interrogating files using the facilities in Neural Planner or using any editor, word processor or spreadsheet that supports text files. Neural Planner can learn from training files, self test using testing files and be interrogated by interrogating files. It can produce spreadsheet or hierarchical output files. It can also be used interactively. |