When you use customized recommendation solutions of PAI-Rec, you can use the data generated by the recall model to generate recommendations on details pages. You can use the data of UserGroupHotRecall and UserGlobalHotRecall to supplement global recommendation data. You can also use the item-to-item (I2I) data and popular item data generated in systems such as Spark and Apache Flink as recall data, as long as the data conforms to the recall data format of the PAI-Rec engine.
Configure the PAI-Rec engine
Recall and filtering logic:
ProductDetailPageEtrecRecall: generates candidate items by using the collaborative filtering algorithm eTREC. These candidate items have the highest priority.
GroupHotRecall: generates top candidate items by categorical features of items. In the following configuration example, the categorical feature category is used.
GlobalHotRecall: generates top candidate items based on the overall information of items. If the candidate items generated by ProductDetailPageEtrecRecall and GroupHotRecall are insufficient, the candidate items generated by GlobalHotRecall are used as supplements.
UserExposureFilter: blocks the items that have been recommended to the same user.
{
"RecallConfs": [
{
"Name": "ProductDetailPageEtrecRecall",
"RecallType": "ItemCollaborativeFilterRecall",
"RecallCount": 50,
"ItemCollaborativeDaoConf": {
"AdapterType": "hologres",
"HologresName": "pairec-holo",
"Item2ItemTable": "i2i_table"
}
},
{
"Name":"GroupHotRecall",
"RecallType": "UserGroupHotRecall",
"RecallCount" :500,
"Triggers": [
{
"TriggerKey": "category"
}
],
"DaoConf":{
"AdapterType": "hologres",
"HologresName": "holo_info",
"HologresTableName": "group_hotness_table"
}
},
{
"Name":"GlobalHotRecall",
"RecallType": "UserGlobalHotRecall",
"RecallCount" :500,
"DaoConf":{
"AdapterType": "hologres",
"HologresName": "holo_info",
"HologresTableName": "global_hotness_table"
}
}
],
"FilterConfs": [
{
"Name": "UserExposureFilter",
"FilterType": "User2ItemExposureFilter",
"MaxItems": 100,
"TimeInterval": 7200,
"WriteLog": true,
"DaoConf": {
"AdapterType": "hologres",
"HologresName": "pairec-holo",
"HologresTableName": "exposure_history"
}
}
],
"SortNames": {
"default": [
"ItemRankScore"
]
},
"FilterNames": {
"default": [
"UniqueFilter",
"UserExposureFilter"
]
},
"AlgoConfs": [],
"HologresConfs": {
"pairec-holo": {
"DSN": "postgres://{AK}:{SK}@{Host}:{Port}/{DB}?sslmode=disable&connect_timeout=1"
}
},
"KafkaConfs": {},
"RedisConfs": {},
"SceneConfs": {
"product_detail_page": {
"default": {
"RecallNames": [
"ProductDetailPageEtrecRecall"
]
}
}
},
"LogConf": {
"RetensionDays": 3,
"DiskSize": 20,
"LogLevel": "INFO"
},
"RankConf": {},
"FeatureConfs": {}
}Configure the engine interface
Request path
/api/rec/feed
Request parameters
Parameter | Description | Type | Required | Example |
uid | The user ID. | string | Yes | "1000079" |
item_id | The item ID. | string | Yes | "9237438" |
size | The number of items to be obtained. | integer | Yes | 10 |
scene_id | The scenario ID. | String | Yes | home_feed |
features | The context features. | json map | No | {"current_detail_item_id":20, "sex":"male"} |
debug | Specifies the logs for debugging. | bool | No | true |
Sample requests
curl -v http://host/api/rec/feed -d '
{"uid":"76295990",
"item_id":"9237438"
"size":1,
"scene_id":"home_feed"}' Sample responses
{
"code":200,
"msg":"success",
"request_id":"e332fe9c-7d99-45a8-a047-bc7ec33d07f6",
"size":1,
"items":[
{
"item_id":"3v5RE7417j7R",
"score":0.0045,
"retrieve_id":"ProductDetailPageEtrecRecall"
}
]
}