jstarcraft example

JStarCraft Example

---

I hope that classmates passing by can give the JStarCraft framework a star, which can be regarded as an encouragement to the author!

Table of contents

• introduce
• characteristic
  • Personalized model
• Install
  • Install JStarCraft Core framework
  • Install JStarCraft AI framework
  • Install JStarCraft RNS engine
  • Packaging the JStarCraft Example project
• use
  • Run the JStarCraft Example project
  • Personalized recommendation and search for movies
  • Automatically adjust models based on behavioral records
• Architecture
• concept
• Example
  • term query
  • range query
  • wildcard query
  • fuzzy query
  • Combined query
• contrast
• Version
• refer to
• protocol
• author
• Acknowledgments

---

introduce

JStarCraft Example is a demonstration project based on JStarCraft RNS engine, Spring Boot framework and public data sets.

The system will automatically adjust the user's recommended content and search content based on the user's behavior records. Users can understand the operation processes of the recommendation system and search system through this project.

It covers two parts: personalized recommendation and personalized search.

---

characteristic

Personalized model

This demonstration project uses the following 8 personalized models:

• AssociationRule
• MostPopular
• Random
• BPR
• ItemKNN
• LDA
• UserKNN
• WRMF

Click to learn more about personalized models

---

Install

In order to focus as much as possible on the demonstration of personalized recommendations and personalized search, the project does not include the deployment of any redundant components (such as MySQL/Redis/Spark/Elasticsearch).

JStarCraft Examlpe requires users to have the following environment:

• JDK 8 or above
• Maven 3

Install JStarCraft-Core framework

git clone https://github.com/HongZhaoHua/jstarcraft-core.git

mvn install -Dmaven.test.skip=true

Install JStarCraft-AI framework

git clone https://github.com/HongZhaoHua/jstarcraft-ai.git

mvn install -Dmaven.test.skip=true

Install JStarCraft-RNS engine

git clone https://github.com/HongZhaoHua/jstarcraft-rns.git

mvn install -Dmaven.test.skip=true

Packaging the JStarCraft-Example project

git clone https://github.com/HongZhaoHua/jstarcraft-example.git

mvn package -Dmaven.test.skip=true

---

use

Run the JStarCraft-Example project

java -jar jstarcraft-example-1.0.jar

Personalized recommendation and search for movies

1. Copy the link http://127.0.0.1:8080/movie.html to the browser.
2. Select user
3. Select model
4. Fill in the query conditions
5. Choose whether to filter (reviewed items)
   • Selecting Yes is equivalent to personalized recommendation
   • Selecting No is equivalent to personalized search
6. Click to personalize

The personalized effect is shown in the figure:

Automatically adjust models based on behavioral records

Users can rate the item by clicking on评价:1 2 3 4 5 corresponding scores, and the system will automatically refresh the model every 5 minutes.

---

Architecture

---

concept

The nature of recommendation and search

---

Example

term query

Supported words and sentences:

• word
• statement (enclosed using double quotes "" )

 Story
"Toy Story"

range query

Supports specifying minimum and maximum values:

• {}angle brackets indicate that the minimum or maximum value is not included
• [] Square brackets indicate that the minimum or maximum value is included

 [1990,2000}

wildcard query

Supports combining wildcards in words or sentences:

• Use ? to match a single character
• Use * to match 0 or more characters

 te?t
te*t

fuzzy query

 test~

Combined query

Supports multiple logical operators:

• Use && to implement intersection operations
• Use || to implement union operations
• Use ! to implement difference set operation
• Use () to implement grouping operations to form more complex logical queries

 1990 || (Toy && Story)

---

contrast

---

Version

---

refer to

---

protocol

JStarCraft Example follows the Apache 2.0 protocol, and all derivative works based on it belong to the author of the derivative work.

---

author

author	Hong Zhaohua
Email	[email protected], [email protected]

---

Acknowledgments

Special thanks to the LibRec team for their support and help in recommendation.

Special thanks to Lu Xugang for his support and help in searching.

---