This is an active window with an embedded KEEL Engine.
This demonstration is intended to demonstrate an application that can monitor and evaluate information that might be available to a brokerage company, a news organization or stock exchange that could uses this type of system. This demonstration uses a KEEL engine to react to transaction information and suggest whether to buy, sell, or hold a specific stock.
This demonstration of a KEEL based solution suggests that it might be valuable to combine a complex set of relationships between stock indicators to make judgment decisions on whether to buy or sell a stock.
The concept being demonstrated suggests that each indicator has a different impact on the final decision. Because Compsim is not an expert in the stock market, the indicators may or may not be the ones that should be chosen. Additionally, there may be many more indicators that should be considered.
This demonstration on Compsim's web site is a Macromedia Flash application. The code to position the scroll bars and the response indicators would be developed by the programmer using common Macromedia tools. The remainder of the Actionscript code (not shown) is created by the KEEL toolkit. It is in the KEEL toolkit, where the relationships between the stock indicators, the setup page and the output presentation are defined. The KEEL toolkit has a menu option to translate the KEEL design to Actionscript as well as Visual Basic, VB .NET, C, Microsoft C#, C++, C++ .NET, Java, Python, and PLC Structured Text. The resulting solution is a very small footprint engine.
This system is based on Compsim's patented and patent pending KEEL technology that covers the toolkit user interface, Compsim's decision-making algorithms, and the resulting KEEL engine architecture.
The snapshot image below shows a high level view of the design of the system from within the KEEL toolkit.
In the KEEL toolkit, the vertical bars indicate decisions or actions. Some of the setup controls have been hidden to simplify the display.
The view above shows how the "Buying by Company Officers" impacts the decision. Similar views show how the other inputs impact the decision. The wires between the decisions define how one decision impacts another. The scroll bars at the bottom indicate inputs (from outside the KEEL engine or internal connections driven by actions internal to the KEEL engine).
This image shows the section of the design associated with establishing the sensitivity of the system to buy / sell / hold transactions.
The following image shows a "dashboard" that can be used to simulate a final user interface to the application from within the KEEL development environment.
The dashboard allows the system to be tested with simulated input data in the units appropriate to the specific application. All data is normalized before it is supplied to the KEEL engine in the final application. While in the design mode, the designer can visualize how one action can impact another. By adjusting the scroll bars at the bottom of the display, the designer can see the exact system response. While this image is probably not readable in this document, it is easy to create in the KEEL toolkit. While in the design mode the user can "see" the design in action and graph the activity.
The opening screen for the StockBuySell application shows five stock indicators for a particular stock:
The scroll bars allow the user to simulate the indicators collected from external sources.
The red text below the scroll bars provides the decision created by the KEEL engine as it combines the inputs. (Hold in this case.)
Below the decision indicator are indications of the strength of the different options. (Buy - 77.20606, Sell - 77.20606, Hold - 80).
Four of the five indicators (Company Officers buying / selling, Company News positive / negative, Industry Trends positive / negative, Unemployment Trends positive / negative) are centered in the middle of the range indicating that in this case, they have no impact on the decision.
The fifth indicator (Short Selling) is set at "None" end of the range, also indicating that it has no impact on the decision.
The results show that the decision is to Hold the present position with the stock.
All of this analysis is based on how the system was set up to operate. The following Setup screen (activated by the Setup menu option) allows the user to adjust the importance of the different indicators in making the buy / sell decision.
This shows the following relative settings (Contribution Adjustments):
Adjusting these values will change the impact of these items in the buy / sell decision.
There are also three sensitivity adjustments.
The settings for these values are best determined from within the KEEL toolkit, but they are provided here so the user can see the impact.
The assumption in this demonstration is that the normal state is to take no action (Hold) and that the external indicators might be used to trigger a buy or a sell. Therefore if the Tuning Base Buy and Tuning Base Sell are closer to the Tuning Base Hold, then the system will be more sensitive to changes in the indicators (more likely to trigger a buy or sell). On the other hand, if the Tuning Base Buy and Tuning Base Sell are further from the Tuning Base Hold, then the system will be less sensitive.
There are two buttons on the main screen that allow the user to get a limited view of the KEEL toolkit screens. These are screen captures and not operational.
This demonstration defines only linear relationships between indicators and a buy / sell decision. There are also no inter-relationships between indicators which would likely be the case in a "real" application. This demonstration is intended to provide a simple suggestion to the type of system that could be created with KEEL technology.
The demonstration applications on Compsim's web site are not intended to be tutorials on the use of the KEEL toolkit. If you are interested in learning more about Knowledge Enhanced Electronic Logic and KEEL technology, please contact Compsim at the phone number listed below.