该文档涉及的组件

ItemCf:UsersPerItem推荐 (ItemCfUsersPerItemRecommBatchOp)

Java 类名:com.alibaba.alink.operator.batch.recommendation.ItemCfUsersPerItemRecommBatchOp

Python 类名:ItemCfUsersPerItemRecommBatchOp

功能介绍

用ItemCF模型 为item推荐user list。

参数说明

名称 中文名称 描述 类型 是否必须? 取值范围 默认值
itemCol Item列列名 Item列列名 String
recommCol 推荐结果列名 推荐结果列名 String
excludeKnown 排除已知的关联 推荐结果中是否排除训练数据中已知的关联 Boolean false
initRecommCol 初始推荐列列名 初始推荐列列名 String 所选列类型为 [M_TABLE, STRING] null
k 推荐TOP数量 推荐TOP数量 Integer 10
reservedCols 算法保留列名 算法保留列 String[] null
numThreads 组件多线程线程个数 组件多线程线程个数 Integer 1

代码示例

Python 代码

from pyalink.alink import *

import pandas as pd

useLocalEnv(1)

df_data = pd.DataFrame([
    [1, 1, 0.6],
    [2, 2, 0.8],
    [2, 3, 0.6],
    [4, 1, 0.6],
    [4, 2, 0.3],
    [4, 3, 0.4],
])

data = BatchOperator.fromDataframe(df_data, schemaStr='user bigint, item bigint, rating double')

model = ItemCfTrainBatchOp()\
    .setUserCol("user")\
    .setItemCol("item")\
    .setRateCol("rating").linkFrom(data);

predictor = ItemCfUsersPerItemRecommBatchOp()\
    .setItemCol("item")\
    .setReservedCols(["item"])\
    .setK(1)\
    .setRecommCol("prediction_result");

predictor.linkFrom(model, data).print()

Java 代码

import org.apache.flink.types.Row;

import com.alibaba.alink.operator.batch.BatchOperator;
import com.alibaba.alink.operator.batch.recommendation.ItemCfTrainBatchOp;
import com.alibaba.alink.operator.batch.recommendation.ItemCfUsersPerItemRecommBatchOp;
import com.alibaba.alink.operator.batch.source.MemSourceBatchOp;
import org.junit.Test;

import java.util.Arrays;
import java.util.List;

public class ItemCfUsersPerItemRecommBatchOpTest {
	@Test
	public void testItemCfUsersPerItemRecommBatchOp() throws Exception {
		List <Row> df_data = Arrays.asList(
			Row.of(1, 1, 0.6),
			Row.of(2, 2, 0.8),
			Row.of(2, 3, 0.6),
			Row.of(4, 1, 0.6),
			Row.of(4, 2, 0.3),
			Row.of(4, 3, 0.4)
		);
		BatchOperator <?> data = new MemSourceBatchOp(df_data, "user int, item int, rating double");
		BatchOperator <?> model = new ItemCfTrainBatchOp()
			.setUserCol("user")
			.setItemCol("item")
			.setRateCol("rating").linkFrom(data);
		BatchOperator <?> predictor = new ItemCfUsersPerItemRecommBatchOp()
			.setItemCol("item")
			.setReservedCols("item")
			.setK(1)
			.setRecommCol("prediction_result");
		predictor.linkFrom(model, data).print();
	}
}

运行结果

item prediction_result
1 {“user”:“[2]”,“score”:“[0.21698238771519462]”}
2 {“user”:“[2]”,“score”:“[0.29215236292253793]”}
3 {“user”:“[2]”,“score”:“[0.38953648389671724]”}
1 {“user”:“[2]”,“score”:“[0.21698238771519462]”}
2 {“user”:“[2]”,“score”:“[0.29215236292253793]”}
3 {“user”:“[2]”,“score”:“[0.38953648389671724]”}