##GM Capstone README

This is the README for the ASU / GM eProject, also known as a Capstone.

The Web Page files will reside under src/main/webapp

The Java files will reside under src/main/java/com/gm (the com/gm is a java standard package naming convention)

GM-Capstone/
├── src/
│...├── main/
│......├── java/
│......│...├── com/
│......│......├── gm/ (The java files will be here in separate packages)
│......├── webapp (HTML/CSS/Javascript files will be here)

│......├── resources
│.........├── META-INF (Contains the persistence.xml file used for defining persistence units)
│.........├── SQL (Contains the database model and SQL files)

• [Maven 3.2.5] (http://maven.apache.org/download.cgi)
• [MySQL Community Server 5.6] (http://dev.mysql.com/downloads/mysql/)
• [JDK 1.6 ] (http://www.oracle.com/technetwork/java/javase/downloads/index.html)
• [Ant 1.9.4] (http://ant.apache.org/bindownload.cgi)
• [Couchdb 1.6.1] (http://couchdb.apache.org/)

1. Uses "mvn" to build, simply use the following command in the root directory:

• mvn clean install

2. TODO: what else we need to do

├── gm/
│...├── twitterstream/
│......├── lib/
│......├── src/
│......├── build.xml
│......├── couchdb_setup.bat │......├── twitter4j.properties │......├── build.properties

1. Database setup script in src/main/resources/sql/master-gm-db-setup.sql must have been run on machine with a MySQL install. In MySQL Workbench:

• File > New Query Tab
• Type "CREATE DATABASE dbname;"
  -- # NOTE: Currently, dbname = testGM
• Query > execute
  -- # OR, click the lightning bolt
• File > Open SQL Script... > master-gm-db-setup.sql
• Query > execute -- # OR, click the lightning bolt

2. Ensure your SQL root password is correct:

• In build.properties input your credentials
• stanbol points to your default enhancement chain described below

3. On machine with couch installed, run couch_setup.bat

• set couch property in build.properties to point to your db

4. twitter4j.properties contains your twitter streaming api key

• obtain an access token from twitter

5. Uses "ant" to build, simply use the following commands in the twitterstream directory:

• ant compile
• ant execute

1. There is no server provided for Apache Stanbol, you must run your own instance of the service.

• export MAVEN_OPTS="-Xmx1024M -XX:MaxPermSize=256M"
• svn co http://svn.apache.org/repos/asf/stanbol/trunk stanbol
• cd stanbol
• mvn clean install
• cd launchers
• java -Xmx1g -jar stable/target/org.apache.stanbol.launchers.stable-1.0.0-SNAPSHOT.jar

2. In a new terminal, go to the following directory in your stanbol directory:

• cd stanbol/data/sentiment/sentiwordnet
• mvn install -DskipTests -PinstallBundle -Dsling=http://localhost:8080/system/console -- This needs to be done each time the instance of stanbol is launched.

3. Add sentiment-wordclassifier and sentiment-summarization to the engines in the default chain

• Go here: http://localhost:8080/enhancer/chain
• Click the configure link next to default, default credentials admin/admin
• Set the engine chain to this order:
• langdetect
• opennlp-sentence
• opennlp-token
• opennlp-pos
• opennlp-ner
• sentiment-wordclassifier
• sentiment-summarization

4. Stanbol's endpoint example

• curl -X POST -H "Accept: application/json" -H "Content-type: text/plain" --data "This is a positive sentence because I like sentences. This is a negative sentence because I hate sentences. I like animals. I hate hats." http://localhost:8080/enhancer