Stream Jess Guide
User Manual:
Open the PDF directly: View PDF 
.
Page Count: 5
| Download | |
| Open PDF In Browser | View PDF |
User Manual StreamJess v0.1 Table of Contents I. Installation ............................................................................................................................................ 3 II. Getting started examples...................................................................................................................... 3 2.1. Stream Generator default settings ............................................................................................... 3 2.2. Jess reasoning ............................................................................................................................... 4 III. User-defined examples ..................................................................................................................... 5 3.1. Change data .................................................................................................................................. 5 3.2. Change the model ......................................................................................................................... 5 3.3. Change the Jess rules .................................................................................................................... 5 References .................................................................................................................................................... 5 I. Installation Download application’s source distribution from the Download section. Unzip the zip file into your local folder. Import the project into your NetBeans. Download a copy of the InWaterSense ontology Protege project including all ontology modules from the Download section. Download and import all the jar libraries into your project including: C-SPARQL1 v0.9.6, Jess2 v7.1p2, Jess Tab3 v1.7 and Protégé4 v3.5. Open main\StreamJess.java and replace the InWaterSense ontology project path with your local copy of the ontology (modify the value of inwsOntologies variable). Moreover, modify the path location of Jess rules to match the one of your local disc i.e. the inwsRules variable value. Make sure you are connected to the Internet for loading correctly the ontology modules and run the application. If everything was correctly loaded the following screenshot should appear: II. Getting started 2.1. Dataset and Stream Generator default settings The default stream generator (INWSRDFStreamTestGenerator.java) produces sensor simulated values on 70 different measurement sites for 11 water quality parameters. Our implemented WQM example includes WFD monitoring of the following water quality parameters: Biochemical Oxygen Demand (BOD5), pH, Arsenic, Benthic Invertebrate Fauna, Chromium III, "Cyanide", Diazinon, Dimethoate, 1 http://streamreasoning.org/resources/c-sparql http://www.jessrules.com/ 3 http://www.jessrules.com/jesswiki/view?JessTab 4 http://protege.stanford.edu/ 2 Fluoride, Total Ammonia and Glyphosate. Furthermore, it identifies the potential sources of pollution inducted by BOD5 and pH moderate values. All values are randomly generated. Two C-SPARQL queries are encoded in the application’s main file StreamJess.java. One for considering observations one by one, which according to Water Framework Directive (WFD) [1] is the case of pH observations and one for average value observations, which include all other water quality observations according to WFD. After correctly loading all the ontology modules and setting up the Jess’s working memory by mapping the required classes as specified by initiateINWSrules.jess file, the user is presented with a dialog for choosing the type of the windows for the average value observations and for individual ones. The streaming of observations starts immediately after choosing the windows type and each second new observation is produced. Of course, this is a modifiable parameter which can be set by changing the milliseconds values in Thread.sleep(1000). Moreover, the user can also register multiple streamers, which can be enabled by setting the value of variable NUM_STREAMS in StreamJess.java. 2.2. Jess reasoning After C-SPARQL processes the first window results, the Jess engine acts to do the inference based on the registered rules. Namely, based on C-SPARQL design principle each query attaches new result formatter. Except for formatting the output, StreamJess uses the function RDFResultsFormatter(r), where r is the instance of the Rete engine, to also do rule-based reasoning. For each new query result, firstly, new assertions are published into the knowledge base and then the Rete engine runs to do the rule-based reasoning. The rules’ output gets printed out on the console as depicted in the following screenshot: III. User-defined examples This section contains information for users willing to write their own examples in StreamJess. 3.1. Change the dataset In case one would like to use different water quality monitoring data, it will be required to stop the streamer(s) and find appropriate middleware for importing real-time stream data and annotate them appropriately based on the InWaterSense ontology pattern. In cases of importing stream data annotated differently from InWaterSense ontology descriptions then he/she should provide their own model, write their own Jess rules and potentially write appropriate Java codes. Of course, in these cases we prefer the users to contact us. 3.2. Change the model If one would like to use other model than InWaterSense ontology, then he/she would have to change the variable path value of inwsOntologies, configure appropriately the stream generator and write appropriate Jess rules. 3.3. Change the Jess rules Using the InWaterSense model, users are open to add other Jess rules for monitoring or finding potential sources of pollution. One can write a Jess rule, save it as a .jess file in the jessRules folder of the project and write a Jess command like follows: r.batch(inwsRules + "xxx.jess"); Place this line right after creating the Rete engine, but not after the running of C-SPARQL engine. References [1] Directive 2000/60/EC of the European Parliament and of the Council of Europe of 23 October 2000 establishing a frame-work for Community action in the Field of water quality O.J. L327/1, 2000.
Source Exif Data:
File Type : PDF File Type Extension : pdf MIME Type : application/pdf PDF Version : 1.5 Linearized : No Page Count : 5 Language : en-US Tagged PDF : Yes Author : none Creator : Microsoft® Word 2010 Create Date : 2016:12:10 21:19:32+01:00 Modify Date : 2016:12:10 21:19:32+01:00 Producer : Microsoft® Word 2010EXIF Metadata provided by EXIF.tools