Home>Programming related>Other source code
Download

JStarCraft RNS

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
  • Install
    • Install JStarCraft Core framework
    • Install JStarCraft AI framework
    • Install JStarCraft RNS engine
  • use
    • Set dependencies
    • Build configurator
    • Training and evaluating models
    • Get model
  • Architecture
  • concept
    • Why information retrieval is needed
    • Similarities and Differences between Search and Recommendation
    • What problems does the JStarCraft RNS engine solve?
    • The difference between Ranking task and Rating task
    • Can the Rating algorithm be used for Ranking problems?
  • Example
    • JStarCraft RNS engine interacts with BeanShell scripts
    • JStarCraft RNS engine interacts with Groovy scripts
    • JStarCraft RNS engine interacts with JS scripts
    • JStarCraft RNS engine interacts with Kotlin scripts
    • JStarCraft RNS engine interacts with Lua scripts
    • JStarCraft RNS engine interacts with Python scripts
    • JStarCraft RNS engine interacts with Ruby scripts
  • contrast
    • Ranking model comparison
    • Scoring model comparison
  • Version
  • refer to
    • Personalized model description
    • Dataset description
  • protocol
  • author
  • Acknowledgments
  • Donate

introduce

JStarCraft RNS is a lightweight engine oriented to the field of information retrieval. It follows the Apache 2.0 protocol.

Focus on solving basic problems in the field of information retrieval: recommendation and search.

Provide recommendation engine design and implementation that meets the requirements of industrial-level scenarios.

Provide search engine design and implementation that meets the requirements of industrial-level scenarios.

characteristic

  • 1. Cross-platform
  • 2. Serial and parallel computing
  • 3.CPU and GPU hardware acceleration
  • 4. Model saving and loading
  • 5. Rich recommendation and search algorithms
  • 6. Rich script support
    • Groovy
    • JS
    • Lua
    • MVEL
    • Python
    • Ruby
  • 7. Rich evaluation indicators
    • Sorting index
    • Rating indicators

Install

JStarCraft RNS 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

use

Set dependencies

  • Set Maven dependencies 
 <dependency>
    <groupId>com.jstarcraft</groupId>
    <artifactId>rns</artifactId>
    <version>1.0</version>
</dependency>
  • Set Gradle dependencies 
compile group : ' com.jstarcraft ' , name : ' rns ' , version : ' 1.0 '

Build configurator 

 Properties keyValues = new Properties ();
keyValues . load ( this . getClass (). getResourceAsStream ( "/data.properties" ));
keyValues . load ( this . getClass (). getResourceAsStream ( "/recommend/benchmark/randomguess-test.properties" ));
Configurator configurator = new Configurator ( keyValues );

Training and evaluating models

  • Build sorting tasks 
 RankingTask task = new RankingTask ( RandomGuessModel . class , configurator );
// 训练与评估排序模型
task . execute ();
  • Build scoring tasks 
 RatingTask task = new RatingTask ( RandomGuessModel . class , configurator );
// 训练与评估评分模型
task . execute ();

Get model 

 // 获取模型
Model model = task . getModel ();

Architecture

concept

Why information retrieval is needed 

随着信息技术和互联网的发展,人们逐渐从信息匮乏(Information Underload)的时代走入了信息过载(Information Overload)的时代.

无论是信息消费者还是信息生产者都遇到了挑战:
* 对于信息消费者,从海量信息中寻找信息,是一件非常困难的事情;
* 对于信息生产者,从海量信息中暴露信息,也是一件非常困难的事情;

信息检索的任务就是联系用户和信息,一方面帮助用户寻找对自己有价值的信息,另一方面帮助信息暴露给对它感兴趣的用户,从而实现信息消费者和信息生产者的双赢.

Similarities and Differences between Search and Recommendation 

从信息检索的角度:
* 搜索和推荐是获取信息的两种主要手段;
* 搜索和推荐是获取信息的两种不同方式;
    * 搜索(Search)是主动明确的;
    * 推荐(Recommend)是被动模糊的;

搜索和推荐是两个互补的工具.

What problems does the JStarCraft-RNS engine solve? 

 JStarCraft-RNS引擎旨在解决推荐与搜索领域的两个核心任务:排序预测(Ranking)和评分预测(Rating).

The difference between Ranking task and Rating task 

根据解决基本问题的不同,将算法与评估指标划分为排序(Ranking)与评分(Rating).

两者之间的根本区别在于目标函数的不同.
通俗点的解释:
Ranking算法基于隐式反馈数据,趋向于拟合用户的排序.(关注度)
Rating算法基于显示反馈数据,趋向于拟合用户的评分.(满意度)

Can the Rating algorithm be used for Ranking problems? 

关键在于具体场景中,关注度与满意度是否保持一致.
通俗点的解释:
人们关注的东西,并不一定是满意的东西.(例如:个人所得税)

Example

JStarCraft-RNS engine interacts with BeanShell scripts

  • Complete example

  • Write a BeanShell script to train and evaluate the model and save it to the Model.bsh file 

 // 构建配置
keyValues = new Properties();
keyValues.load(loader.getResourceAsStream("data.properties"));
keyValues.load(loader.getResourceAsStream("model/benchmark/randomguess-test.properties"));
configurator = new Configurator(keyValues);

// 此对象会返回给Java程序
_data = new HashMap();

// 构建排序任务
task = new RankingTask(RandomGuessModel.class, configurator);
// 训练与评估模型并获取排序指标
measures = task.execute();
_data.put("precision", measures.get(PrecisionEvaluator.class));
_data.put("recall", measures.get(RecallEvaluator.class));

// 构建评分任务
task = new RatingTask(RandomGuessModel.class, configurator);
// 训练与评估模型并获取评分指标
measures = task.execute();
_data.put("mae", measures.get(MAEEvaluator.class));
_data.put("mse", measures.get(MSEEvaluator.class));

_data;
  • Load and execute BeanShell scripts from the Model.bsh file using the JStarCraft framework 
 // 获取BeanShell脚本
File file = new File ( ScriptTestCase . class . getResource ( "Model.bsh" ). toURI ());
String script = FileUtils . readFileToString ( file , StringUtility . CHARSET );

// 设置BeanShell脚本使用到的Java类
ScriptContext context = new ScriptContext ();
context . useClasses ( Properties . class , Assert . class );
context . useClass ( "Configurator" , MapConfigurator . class );
context . useClasses ( "com.jstarcraft.ai.evaluate" );
context . useClasses ( "com.jstarcraft.rns.task" );
context . useClasses ( "com.jstarcraft.rns.model.benchmark" );
// 设置BeanShell脚本使用到的Java变量
ScriptScope scope = new ScriptScope ();
scope . createAttribute ( "loader" , loader );

// 执行BeanShell脚本
ScriptExpression expression = new GroovyExpression ( context , scope , script );
Map < String , Float > data = expression . doWith ( Map . class );
Assert . assertEquals ( 0.005825241F , data . get ( "precision" ), 0F );
Assert . assertEquals ( 0.011579763F , data . get ( "recall" ), 0F );
Assert . assertEquals ( 1.2708743F , data . get ( "mae" ), 0F );
Assert . assertEquals ( 2.425075F , data . get ( "mse" ), 0F );

JStarCraft-RNS engine interacts with Groovy scripts

  • Complete example

  • Write Groovy scripts to train and evaluate the model and save it to the Model.groovy file 

 // 构建配置
def keyValues = new Properties ();
keyValues . load(loader . getResourceAsStream( " data.properties " ));
keyValues . load(loader . getResourceAsStream( " recommend/benchmark/randomguess-test.properties " ));
def configurator = new Configurator (keyValues);

// 此对象会返回给Java程序
def _data = [:];

// 构建排序任务
task = new RankingTask ( RandomGuessModel . class, configurator);
// 训练与评估模型并获取排序指标
measures = task . execute();
_data . precision = measures . get( PrecisionEvaluator . class);
_data . recall = measures . get( RecallEvaluator . class);

// 构建评分任务
task = new RatingTask ( RandomGuessModel . class, configurator);
// 训练与评估模型并获取评分指标
measures = task . execute();
_data . mae = measures . get( MAEEvaluator . class);
_data . mse = measures . get( MSEEvaluator . class);

_data;
  • Load and execute Groovy scripts from Model.groovy files using the JStarCraft framework 
 // 获取Groovy脚本
File file = new File ( ScriptTestCase . class . getResource ( "Model.groovy" ). toURI ());
String script = FileUtils . readFileToString ( file , StringUtility . CHARSET );

// 设置Groovy脚本使用到的Java类
ScriptContext context = new ScriptContext ();
context . useClasses ( Properties . class , Assert . class );
context . useClass ( "Configurator" , MapConfigurator . class );
context . useClasses ( "com.jstarcraft.ai.evaluate" );
context . useClasses ( "com.jstarcraft.rns.task" );
context . useClasses ( "com.jstarcraft.rns.model.benchmark" );
// 设置Groovy脚本使用到的Java变量
ScriptScope scope = new ScriptScope ();
scope . createAttribute ( "loader" , loader );

// 执行Groovy脚本
ScriptExpression expression = new GroovyExpression ( context , scope , script );
Map < String , Float > data = expression . doWith ( Map . class );

JStarCraft-RNS engine interacts with JS scripts

  • Complete example

  • Write JS scripts to train and evaluate the model and save it to the Model.js file 

 // 构建配置
var keyValues = new Properties ( ) ;
keyValues . load ( loader . getResourceAsStream ( "data.properties" ) ) ;
keyValues . load ( loader . getResourceAsStream ( "recommend/benchmark/randomguess-test.properties" ) ) ;
var configurator = new Configurator ( [ keyValues ] ) ;

// 此对象会返回给Java程序
var _data = { } ;

// 构建排序任务
task = new RankingTask ( RandomGuessModel . class , configurator ) ;
// 训练与评估模型并获取排序指标
measures = task . execute ( ) ;
_data [ 'precision' ] = measures . get ( PrecisionEvaluator . class ) ;
_data [ 'recall' ] = measures . get ( RecallEvaluator . class ) ;

// 构建评分任务
task = new RatingTask ( RandomGuessModel . class , configurator ) ;
// 训练与评估模型并获取评分指标
measures = task . execute ( ) ;
_data [ 'mae' ] = measures . get ( MAEEvaluator . class ) ;
_data [ 'mse' ] = measures . get ( MSEEvaluator . class ) ;

_data ;
  • Load and execute JS scripts from the Model.js file using the JStarCraft framework 
 // 获取JS脚本
File file = new File ( ScriptTestCase . class . getResource ( "Model.js" ). toURI ());
String script = FileUtils . readFileToString ( file , StringUtility . CHARSET );

// 设置JS脚本使用到的Java类
ScriptContext context = new ScriptContext ();
context . useClasses ( Properties . class , Assert . class );
context . useClass ( "Configurator" , MapConfigurator . class );
context . useClasses ( "com.jstarcraft.ai.evaluate" );
context . useClasses ( "com.jstarcraft.rns.task" );
context . useClasses ( "com.jstarcraft.rns.model.benchmark" );
// 设置JS脚本使用到的Java变量
ScriptScope scope = new ScriptScope ();
scope . createAttribute ( "loader" , loader );

// 执行JS脚本
ScriptExpression expression = new JsExpression ( context , scope , script );
Map < String , Float > data = expression . doWith ( Map . class );

JStarCraft-RNS engine interacts with Kotlin scripts

  • Complete example

  • Write a Kotlin script to train and evaluate the model and save it to the Model.kt file 

 // 构建配置
var keyValues = Properties ( ) ;
var loader = bindings [ "loader" ] as ClassLoader ;
keyValues . load ( loader . getResourceAsStream ( "data.properties" ) ) ;
keyValues . load ( loader . getResourceAsStream ( "model/benchmark/randomguess-test.properties" ) ) ;
var option = Option ( keyValues ) ;

// 此对象会返回给Java程序
var _data = mutableMapOf < String , Float > ( ) ;

// 构建排序任务
var rankingTask = RankingTask ( RandomGuessModel :: class . java , option ) ;
// 训练与评估模型并获取排序指标
val rankingMeasures = rankingTask . execute ( ) ;
_data [ "precision" ] = rankingMeasures . getFloat ( PrecisionEvaluator :: class . java ) ;
_data [ "recall" ] = rankingMeasures . getFloat ( RecallEvaluator :: class . java ) ;

// 构建评分任务
var ratingTask = RatingTask ( RandomGuessModel :: class . java , option ) ;
// 训练与评估模型并获取评分指标
var ratingMeasures = ratingTask . execute ( ) ;
_data [ "mae" ] = ratingMeasures . getFloat ( MAEEvaluator :: class . java ) ;
_data [ "mse" ] = ratingMeasures . getFloat ( MSEEvaluator :: class . java ) ;

_data ;
  • Load and execute Kotlin scripts from Model.kt files using the JStarCraft framework 
 // 获取Kotlin脚本
File file = new File ( ScriptTestCase . class . getResource ( "Model.kt" ). toURI ());
String script = FileUtils . readFileToString ( file , StringUtility . CHARSET );

// 设置Kotlin脚本使用到的Java类
ScriptContext context = new ScriptContext ();
context . useClasses ( Properties . class , Assert . class );
context . useClass ( "Option" , MapOption . class );
context . useClasses ( "com.jstarcraft.ai.evaluate" );
context . useClasses ( "com.jstarcraft.rns.task" );
context . useClasses ( "com.jstarcraft.rns.model.benchmark" );
// 设置Kotlin脚本使用到的Java变量
ScriptScope scope = new ScriptScope ();
scope . createAttribute ( "loader" , loader );

// 执行Kotlin脚本
ScriptExpression expression = new KotlinExpression ( context , scope , script );
Map < String , Float > data = expression . doWith ( Map . class );

JStarCraft-RNS engine interacts with Lua scripts

  • Complete example

  • Write Lua scripts to train and evaluate the model and save it to the Model.lua file 

 -- 构建配置
local keyValues = Properties . new ();
keyValues : load ( loader : getResourceAsStream ( " data.properties " ));

keyValues : load ( loader : getResourceAsStream ( " recommend/benchmark/randomguess-test.properties " ));
local configurator = Configurator . new ({ keyValues });

-- 此对象会返回给Java程序
local _data = {};

-- 构建排序任务
task = RankingTask . new ( RandomGuessModel , configurator );
-- 训练与评估模型并获取排序指标
measures = task : execute ();
_data [ " precision " ] = measures : get ( PrecisionEvaluator );
_data [ " recall " ] = measures : get ( RecallEvaluator );

-- 构建评分任务
task = RatingTask . new ( RandomGuessModel , configurator );
-- 训练与评估模型并获取评分指标
measures = task : execute ();
_data [ " mae " ] = measures : get ( MAEEvaluator );
_data [ " mse " ] = measures : get ( MSEEvaluator );

return _data ;
  • Load and execute Lua scripts from the Model.lua file using the JStarCraft framework 
 // 获取Lua脚本
File file = new File ( ScriptTestCase . class . getResource ( "Model.lua" ). toURI ());
String script = FileUtils . readFileToString ( file , StringUtility . CHARSET );

// 设置Lua脚本使用到的Java类
ScriptContext context = new ScriptContext ();
context . useClasses ( Properties . class , Assert . class );
context . useClass ( "Configurator" , MapConfigurator . class );
context . useClasses ( "com.jstarcraft.ai.evaluate" );
context . useClasses ( "com.jstarcraft.rns.task" );
context . useClasses ( "com.jstarcraft.rns.model.benchmark" );
// 设置Lua脚本使用到的Java变量
ScriptScope scope = new ScriptScope ();
scope . createAttribute ( "loader" , loader );

// 执行Lua脚本
ScriptExpression expression = new LuaExpression ( context , scope , script );
LuaTable data = expression . doWith ( LuaTable . class );

JStarCraft-RNS engine interacts with Python scripts

  • Complete example

  • Write Python scripts to train and evaluate the model and save it to the Model.py file 

 # 构建配置
keyValues = Properties ()
keyValues . load ( loader . getResourceAsStream ( "data.properties" ))
keyValues . load ( loader . getResourceAsStream ( "recommend/benchmark/randomguess-test.properties" ))
configurator = Configurator ([ keyValues ])

# 此对象会返回给Java程序
_data = {}

# 构建排序任务
task = RankingTask ( RandomGuessModel , configurator )
# 训练与评估模型并获取排序指标
measures = task . execute ()
_data [ 'precision' ] = measures . get ( PrecisionEvaluator )
_data [ 'recall' ] = measures . get ( RecallEvaluator )

# 构建评分任务
task = RatingTask ( RandomGuessModel , configurator )
# 训练与评估模型并获取评分指标
measures = task . execute ()
_data [ 'mae' ] = measures . get ( MAEEvaluator )
_data [ 'mse' ] = measures . get ( MSEEvaluator )
  • Load and execute Python scripts from Model.py files using the JStarCraft framework 
 // 设置Python环境变量
System . setProperty ( "python.console.encoding" , StringUtility . CHARSET . name ());

// 获取Python脚本
File file = new File ( PythonTestCase . class . getResource ( "Model.py" ). toURI ());
String script = FileUtils . readFileToString ( file , StringUtility . CHARSET );

// 设置Python脚本使用到的Java类
ScriptContext context = new ScriptContext ();
context . useClasses ( Properties . class , Assert . class );
context . useClass ( "Configurator" , MapConfigurator . class );
context . useClasses ( "com.jstarcraft.ai.evaluate" );
context . useClasses ( "com.jstarcraft.rns.task" );
context . useClasses ( "com.jstarcraft.rns.model.benchmark" );
// 设置Python脚本使用到的Java变量
ScriptScope scope = new ScriptScope ();
scope . createAttribute ( "loader" , loader );

// 执行Python脚本
ScriptExpression expression = new PythonExpression ( context , scope , script );
Map < String , Double > data = expression . doWith ( Map . class );

JStarCraft-Ruby

  • Complete example

  • Write a Ruby script to train and evaluate the model and save it to the Model.rb file 

 # 构建配置
keyValues = Properties . new ( )
keyValues . load ( $loader . getResourceAsStream ( "data.properties" ) )
keyValues . load ( $loader . getResourceAsStream ( "model/benchmark/randomguess-test.properties" ) )
configurator = Configurator . new ( keyValues )

# 此对象会返回给Java程序
_data = Hash . new ( )

# 构建排序任务
task = RankingTask . new ( RandomGuessModel . java_class , configurator )
# 训练与评估模型并获取排序指标
measures = task . execute ( )
_data [ 'precision' ] = measures . get ( PrecisionEvaluator . java_class )
_data [ 'recall' ] = measures . get ( RecallEvaluator . java_class )

# 构建评分任务
task = RatingTask . new ( RandomGuessModel . java_class , configurator )
# 训练与评估模型并获取评分指标
measures = task . execute ( )
_data [ 'mae' ] = measures . get ( MAEEvaluator . java_class )
_data [ 'mse' ] = measures . get ( MSEEvaluator . java_class )

_data ;
  • Load and execute Ruby scripts from the Model.rb file using the JStarCraft framework 
 // 获取Ruby脚本
File file = new File ( ScriptTestCase . class . getResource ( "Model.rb" ). toURI ());
String script = FileUtils . readFileToString ( file , StringUtility . CHARSET );

// 设置Ruby脚本使用到的Java类
ScriptContext context = new ScriptContext ();
context . useClasses ( Properties . class , Assert . class );
context . useClass ( "Configurator" , MapConfigurator . class );
context . useClasses ( "com.jstarcraft.ai.evaluate" );
context . useClasses ( "com.jstarcraft.rns.task" );
context . useClasses ( "com.jstarcraft.rns.model.benchmark" );
// 设置Ruby脚本使用到的Java变量
ScriptScope scope = new ScriptScope ();
scope . createAttribute ( "loader" , loader );

// 执行Ruby脚本
ScriptExpression expression = new RubyExpression ( context , scope , script );
Map < String , Double > data = expression . doWith ( Map . class );
Assert . assertEquals ( 0.005825241096317768D , data . get ( "precision" ), 0D );
Assert . assertEquals ( 0.011579763144254684D , data . get ( "recall" ), 0D );
Assert . assertEquals ( 1.270874261856079D , data . get ( "mae" ), 0D );
Assert . assertEquals ( 2.425075054168701D , data . get ( "mse" ), 0D );

contrast

Ranking model comparison

  • Baseline model
name Dataset training (milliseconds) Prediction (milliseconds) AUC MAP MRR NDCG Novelty Precision Recall
MostPopular filmtrust 43 273 0.92080 0.41246 0.57196 0.51583 11.79295 0.33230 0.62385
RandomGuess filmtrust 38 391 0.51922 0.00627 0.02170 0.01121 91.94900 0.00550 0.01262
  • collaborative model
name Dataset training (milliseconds) Prediction (milliseconds) AUC MAP MRR NDCG Novelty Precision Recall
AHr filmtrust 12448 253 0.89324 0.38967 0.53990 0.48338 21.13004 0.32295 0.56864
AspectRanking filmtrust 177 58 0.85130 0.15498 0.42480 0.26012 37.36273 0.13302 0.31292
BHFreeRanking filmtrust 5720 4257 0.92080 0.41316 0.57231 0.51662 11.79567 0.33276 0.62500
BPR filmtrust 4228 137 0.89390 0.39886 0.54790 0.49180 21.46738 0.32268 0.57623
BUCMRanking filmtrust 2111 1343 0.90782 0.39794 0.55776 0.49651 13.08073 0.32407 0.59141
CDAE filmtrust 89280 376 0.91880 0.40759 0.56855 0.51089 11.82466 0.33051 0.61967
CliMF filmtrust 48429 140 0.88293 0.37395 0.52407 0.46572 19.38964 0.32049 0.54605
DeepFM filmtrust 69264 99 0.91679 0.40580 0.56995 0.50985 11.90242 0.32719 0.61426
EALS filmtrust 850 185 0.86132 0.31263 0.45680 0.39475 20.08964 0.27381 0.46271
FISMAUC filmtrust 2338 663 0.91216 0.40032 0.55730 0.50114 12.07469 0.32845 0.60294
FISMRMSE filmtrust 4030 729 0.91482 0.40795 0.56470 0.50920 11.91234 0.33044 0.61107
GBPR filmtrust 14827 150 0.92113 0.41003 0.57144 0.51464 11.87609 0.33090 0.62512
HMM game 38697 11223 0.80559 0.18156 0.37516 0.25803 16.01041 0.14572 0.22810
ItemBigram filmtrust 12492 61 0.88807 0.33520 0.46870 0.42854 17.11172 0.29191 0.53308
ItemKNNRanking filmtrust 2683 250 0.87438 0.33375 0.46951 0.41767 20.23449 0.28581 0.49248
LDA filmtrust 696 161 0.91980 0.41758 0.58130 0.52003 12.31348 0.33336 0.62274
LambdaFMStatic game 25052 27078 0.87064 0.27294 0.43640 0.34794 16.47330 0.13941 0.35696
LambdaFMWeight game 25232 28156 0.87339 0.27333 0.43720 0.34728 14.71413 0.13742 0.35252
LambdaFMDynamic game 74218 27921 0.87380 0.27288 0.43648 0.34706 13.50578 0.13822 0.35132
ListwiseMF filmtrust 714 161 0.90820 0.40511 0.56619 0.50521 15.53665 0.32944 0.60092
PLSA filmtrust 1027 116 0.89950 0.41217 0.57187 0.50597 16.01080 0.32401 0.58557
RankALS filmtrust 3285 182 0.85901 0.29255 0.51014 0.38871 25.27197 0.22931 0.42509
RankCD product 1442 8905 0.56271 0.01253 0.04618 0.02682 55.42019 0.01548 0.03520
RankSGD filmtrust 309 113 0.80388 0.23587 0.42290 0.32081 42.83305 0.19363 0.35374
RankVFCD product 54273 6524 0.58022 0.01784 0.06181 0.03664 62.95810 0.01980 0.04852
SLIM filmtrust 62434 91 0.91849 0.44851 0.61083 0.54557 16.67990 0.34019 0.63021
UserKNNRanking filmtrust 1154 229 0.90752 0.41616 0.57525 0.51393 12.90921 0.32891 0.60152
VBPR product 184473 15304 0.54336 0.00920 0.03522 0.01883 45.05101 0.01037 0.02266
WBPR filmtrust 20705 183 0.78072 0.24647 0.33373 0.30442 17.18609 0.25000 0.35516
WRMF filmtrust 482 158 0.90616 0.43278 0.58284 0.52480 15.17956 0.32918 0.60780
RankGeoFM FourSquare 368436 1093 0.72708 0.05485 0.24012 0.11057 37.50040 0.07866 0.08640
SBPR filmtrust 41481 247 0.91010 0.41189 0.56480 0.50726 15.67905 0.32440 0.59699
  • content model
name Dataset training (milliseconds) Prediction (milliseconds) AUC MAP MRR NDCG Novelty Precision Recall
EFMRanking dc_dense 2066 2276 0.61271 0.01611 0.04631 0.04045 53.26140 0.02387 0.07357
TFIDF musical_instruments 942 1085 0.52756 0.01067 0.01917 0.01773 72.71228 0.00588 0.03103

Scoring model comparison

  • Baseline model
name Dataset training (milliseconds) Prediction (milliseconds) MAE MPE MSE
ConstantGuess filmtrust 137 45 1.05608 1.00000 1.42309
GlobalAverage filmtrust 60 13 0.71977 0.77908 0.85199
ItemAverage filmtrust 59 12 0.72968 0.97242 0.86413
ItemCluster filmtrust 471 41 0.71976 0.77908 0.85198
RandomGuess filmtrust 38 8 1.28622 0.99597 2.47927
UserAverage filmtrust 35 9 0.64618 0.97242 0.70172
UserCluster filmtrust 326 45 0.71977 0.77908 0.85199
  • collaborative model
name Dataset training (milliseconds) Prediction (milliseconds) MAE MPE MSE
AspectRating filmtrust 220 5 0.65754 0.97918 0.71809
ASVDPlusPlus filmtrust 5631 8 0.71975 0.77921 0.85196
BiasedMF filmtrust 92 6 0.63157 0.98387 0.66220
BHFreeRating filmtrust 6667 76 0.71974 0.77908 0.85198
BPMF filmtrust 25942 52 0.66504 0.98465 0.70210
BUCMRating filmtrust 1843 30 0.64834 0.99102 0.67992
CCD product 15715 9 0.96670 0.93947 1.62145
FFM filmtrust 5422 6 0.63446 0.98413 0.66682
FMALS filmtrust 1854 5 0.64788 0.96032 0.73636
fmsgd filmtrust 3496 10 0.63452 0.98426 0.66710
GPLSA filmtrust 2567 7 0.67311 0.98972 0.79883
IRRG filmtrust 40284 6 0.64766 0.98777 0.73700
ItemKNNRating filmtrust 2052 27 0.62341 0.95394 0.67312
LDCC filmtrust 8650 84 0.66383 0.99284 0.70666
LLORMA filmtrust 16618 82 0.64930 0.96591 0.76067
MFALS filmtrust 2944 5 0.82939 0.94549 1.30547
NMF filmtrust 1198 8 0.67661 0.96604 0.83493
PMF filmtrust 215 7 0.72959 0.98165 0.99948
RBM filmtrust 19551 270 0.74484 0.98504 0.88968
RFRec filmtrust 16330 54 0.64008 0.97112 0.69390
SVDPlusPlus filmtrust 452 26 0.65248 0.99141 0.68289
URP filmtrust 1514 25 0.64207 0.99128 0.67122
UserKNNRating filmtrust 1121 135 0.63933 0.94640 0.69280
RSTE filmtrust 4052 10 0.64303 0.99206 0.67777
SocialMF filmtrust 918 13 0.64668 0.98881 0.68228
SoRec filmtrust 1048 10 0.64305 0.99232 0.67776
SoReg filmtrust 635 8 0.65943 0.96734 0.72760
TimeSVD filmtrust 11545 36 0.68954 0.93326 0.87783
TrustMF filmtrust 2038 7 0.63787 0.98985 0.69017
TrustSVD filmtrust 12465 twenty two 0.61984 0.98933 0.63875
AssociationRule filmtrust 2628 195 0.90853 0.41801 0.57777
PersonalityDiagnosis filmtrust 45 642 0.72964 0.76620 1.03071
PRankD filmtrust 3321 170 0.74472 0.22894 0.32406
SlopeOne filmtrust 135 28 0.63788 0.96175 0.71057
  • content model
name Dataset training (milliseconds) Prediction (milliseconds) MAE MPE MSE
EFMRating dc_dense 659 8 0.61546 0.85364 0.78279
HFT musical_instruments 162753 13 0.64272 0.94886 0.81393
TopicMFAT musical_instruments 6907 7 0.61896 0.98734 0.72545
TopicMFMT musical_instruments 6323 7 0.61896 0.98734 0.72545

Version

refer to

Personalized model description

  • Baseline model
name question Description/thesis
RandomGuess Ranking Rating random guess
MostPopular Ranking most popular
ConstantGuess Rating constant guess
GlobalAverage Rating global average
ItemAverage Rating Item average
ItemCluster Rating Item clustering
UserAverage Rating average user
UserCluster Rating User clustering
  • collaborative model
name question Description/thesis
AspectModel Ranking Rating Latent class models for collaborative filtering
BHFree Ranking Rating Balancing Prediction and Recommendation Accuracy: Hierarchical Latent Factors for Preference Data
BUCM Ranking Rating Modeling Item Selection and Relevance for Accurate Recommendations
ItemKNN Ranking Rating Item-based collaborative filtering
UserKNN Ranking Rating User-based collaborative filtering
AHr Ranking Improving pairwise learning for item recommendation from implicit feedback
BPR Ranking BPR: Bayesian Personalized Ranking from Implicit Feedback
CliMF Ranking CLiMF: learning to maximize reciprocal rank with collaborative less-is-more filtering
EALS Ranking Collaborative filtering for implicit feedback dataset
FISM Ranking FISM: Factored Item Similarity Models for Top-N Recommender Systems
GBPR Ranking GBPR: Group Preference Based Bayesian Personalized Ranking for One-Class Collaborative Filtering
HMMForCF Ranking A Hidden Markov Model Purpose: A class for the model, including parameters
ItemBigram Ranking Topic Modeling: Beyond Bag-of-Words
LambdaFM Ranking LambdaFM: Learning Optimal Ranking with Factorization Machines Using Lambda Surrogates
LDA Ranking Latent Dirichlet Allocation for implicit feedback
ListwiseMF Ranking List-wise learning to rank with matrix factorization for collaborative filtering
PLSA Ranking Latent semantic models for collaborative filtering
RankALS Ranking Alternating Least Squares for Personalized Ranking
RankSGD Ranking Collaborative Filtering Ensemble for Ranking
SLIM Ranking SLIM: Sparse Linear Methods for Top-N Recommender Systems
WBPR Ranking Bayesian Personalized Ranking for Non-Uniformly Sampled Items
WRMF Ranking Collaborative filtering for implicit feedback datasets
Rank-GeoFM Ranking Rank-GeoFM: A ranking based geographical factorization method for point of interest recommendation
SBPR Ranking Leveraging Social Connections to Improve Personalized Ranking for Collaborative Filtering
AssociationRule Ranking A Recommendation Algorithm Using Multi-Level Association Rules
PRankD Ranking Personalized ranking with diversity
AsymmetricSVD++ Rating Factorization Meets the Neighborhood: a Multifaceted Collaborative Filtering Model
AutoRec Rating AutoRec: Autoencoders Meet Collaborative Filtering
BPMF Rating Bayesian Probabilistic Matrix Factorization using Markov Chain Monte Carlo
CCD Rating Large-Scale Parallel Collaborative Filtering for the Netflix Prize
FFM Rating Field Aware Factorization Machines for CTR Prediction
GPLSA Rating Collaborative Filtering via Gaussian Probabilistic Latent Semantic Analysis
IRRG Rating Exploiting Implicit Item Relationships for Recommender Systems
MFALS Rating Large-Scale Parallel Collaborative Filtering for the Netflix Prize
NMF Rating Algorithms for Non-negative Matrix Factorization
PMF Rating PMF: Probabilistic Matrix Factorization
RBM Rating Restricted Boltzman Machines for Collaborative Filtering
RF-Rec Rating RF-Rec: Fast and Accurate Computation of Recommendations based on Rating Frequencies
SVD++ Rating Factorization Meets the Neighborhood: a Multifaceted Collaborative Filtering Model
URP Rating User Rating Profile: a LDA model for rating prediction
RSTE Rating Learning to Recommend with Social Trust Ensemble
SocialMF Rating A matrix factorization technique with trust propagation for recommendation in social networks
SoRec Rating SoRec: Social recommendation using probabilistic matrix factorization
SoReg Rating Recommender systems with social regularization
TimeSVD++ Rating Collaborative Filtering with Temporal Dynamics
TrustMF Rating Social Collaborative Filtering by Trust
TrustSVD Rating TrustSVD: Collaborative Filtering with Both the Explicit and Implicit Influence of User Trust and of Item Ratings
PersonalityDiagnosis Rating A brief introduction to Personality Diagnosis
SlopeOne Rating Slope One Predictors for Online Rating-Based Collaborative Filtering
  • content model
name question Description/thesis
EFM Ranking Rating Explicit factor models for explainable recommendation based on phrase-level sentiment analysis
TF-IDF Ranking Term frequency-inverse document frequency
HFT Rating Hidden factors and hidden topics: understanding rating dimensions with review text
TopicMF Rating TopicMF: Simultaneously Exploiting Ratings and Reviews for Recommendation

Dataset description

  • Amazon Dataset
  • Bibsonomy Dataset
  • BookCrossing Dataset
  • Ciao Dataset
  • Douban Dataset
  • Eachmovie Dataset
  • Epinions Dataset
  • Foursquare Dataset
  • Goodbooks Dataset
  • Gowalla Dataset
  • HetRec2011 Dataset
  • Jest Joker Dataset
  • Large Movie Review Dataset
  • MovieLens Dataset
  • Newsgroups Dataset
  • Stanford Large Network Dataset
  • Serendipity 2018 Dataset
  • Wikilens Dataset
  • Yelp Dataset
  • Yongfeng Zhang Dataset

protocol

JStarCraft RNS 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 and the recommendation system QQ group (274750470) for their support and help in recommendation.

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

Donate

Special thanks to Hangzhou Chengzi Information Technology Co., Ltd. for donating to the JStarCraft project.

Expand
Additional Information
Related Applications
Recommended for You
Related Information All