WO2023109025A1 - Procédé de traitement d'informations de livraison et procédé et appareil d'entraînement d'un modèle de prédiction de ressources - Google Patents

Procédé de traitement d'informations de livraison et procédé et appareil d'entraînement d'un modèle de prédiction de ressources Download PDF

Info

Publication number
WO2023109025A1
WO2023109025A1 PCT/CN2022/096373 CN2022096373W WO2023109025A1 WO 2023109025 A1 WO2023109025 A1 WO 2023109025A1 CN 2022096373 W CN2022096373 W CN 2022096373W WO 2023109025 A1 WO2023109025 A1 WO 2023109025A1
Authority
WO
WIPO (PCT)
Prior art keywords
delivery
information
target
resource
historical
Prior art date
Application number
PCT/CN2022/096373
Other languages
English (en)
Chinese (zh)
Inventor
张弛
郭远
李怀宇
谢淼
林子钏
杨森
刘霁
Original Assignee
北京达佳互联信息技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 北京达佳互联信息技术有限公司 filed Critical 北京达佳互联信息技术有限公司
Publication of WO2023109025A1 publication Critical patent/WO2023109025A1/fr

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90Details of database functions independent of the retrieved data types
    • G06F16/95Retrieval from the web
    • G06F16/953Querying, e.g. by the use of web search engines
    • G06F16/9535Search customisation based on user profiles and personalisation

Definitions

  • the information delivery platform In the information delivery system, new delivery information is continuously uploaded to the system and waits for delivery. In order to quickly identify delivery information with great potential from a large number of newly uploaded delivery information, the information delivery platform generally allocates corresponding cold-start resources to the newly uploaded delivery information, so that they can obtain greater delivery opportunities.
  • the first prediction unit is configured to input the initial state feature information of the target delivery information in the current delivery cycle into the conditional variational self-encoding network to perform resource prediction and obtain the first resource;
  • a computer-readable storage medium When the instructions in the computer-readable storage medium are executed by the processor of the server, the server can execute the method for processing delivery information as described above. Or resource prediction model training method.
  • Fig. 5 is a flow chart showing a method for calculating placement revenue according to an exemplary embodiment.
  • FIG. 7 shows a method for training a resource prediction model, which may include steps S710 to S750.
  • conditional variational self-encoding network can be an independent encoding network.
  • the output of the corresponding independent encoding network includes probability distribution information and encoding of historical resources Information Two pieces of information.
  • the initial state feature information in the last delivery cycle also includes delivery setting information and category information of the target delivery information; the delivery setting information is used to set multiple target delivery information to be delivered Sort;
  • the first updating unit includes: a first generating unit configured to generate the target delivery information at the beginning of the current delivery cycle based on the delivery setting information, the category information, and the updated historical delivery results. Initial state feature information.
  • the actual resource determining unit is configured to determine the actual resource allocated for the target delivery information in the current delivery period based on the normalization coefficient and the preset resource amount;
  • the first sorting unit includes: a second sorting unit configured to, based on the delivery setting information of the items of information to be delivered and the actual resources of the items of information to be delivered, sort the The information to be delivered is sorted to obtain the sorting result.
  • the first training unit 1420 is configured to train a preset conditional variational autoencoder network based on the initial state feature information and the historical resources to obtain a target conditional variational autoencoder network.
  • the third training unit includes:
  • a first loss function determining unit configured to obtain a first loss function based on the first loss component and the second loss component

Landscapes

  • Engineering & Computer Science (AREA)
  • Databases & Information Systems (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

La présente divulgation relève du domaine technique du traitement d'informations. Elle concerne un procédé de traitement d'informations de livraison, ainsi qu'un procédé et un appareil d'entraînement d'un modèle de prédiction de ressources. Le procédé comprend les étapes consistant à : déterminer des informations sur des caractéristiques d'un état initial d'informations de livraison cibles au cours d'une période de livraison actuelle ; obtenir un modèle de prédiction de ressources, le modèle de prédiction de ressources contenant un réseau d'autocodeur variationnel conditionnel et un réseau d'exécution de prédiction ; entrer les informations sur les caractéristiques de l'état initial dans le réseau d'autocodeur variationnel conditionnel en vue d'une prédiction de ressource de façon à obtenir une première ressource ; entrer les informations sur les caractéristiques de l'état initial et la première ressource dans le réseau d'exécution de prédiction en vue d'une prédiction de ressource de façon à obtenir une seconde ressource ; et, sur la base des première et seconde ressources, obtenir une ressource cible correspondant aux informations de livraison cibles, la ressource cible étant une ressource de prédiction qui permet à des recettes de livraison des informations de livraison cibles au cours de la période de livraison actuelle de correspondre à des recettes de livraison cibles.
PCT/CN2022/096373 2021-12-15 2022-05-31 Procédé de traitement d'informations de livraison et procédé et appareil d'entraînement d'un modèle de prédiction de ressources WO2023109025A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202111529876.1A CN113918826B (zh) 2021-12-15 2021-12-15 投放信息处理方法、资源预测模型训练方法及装置
CN202111529876.1 2021-12-15

Publications (1)

Publication Number Publication Date
WO2023109025A1 true WO2023109025A1 (fr) 2023-06-22

Family

ID=79248943

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/096373 WO2023109025A1 (fr) 2021-12-15 2022-05-31 Procédé de traitement d'informations de livraison et procédé et appareil d'entraînement d'un modèle de prédiction de ressources

Country Status (2)

Country Link
CN (1) CN113918826B (fr)
WO (1) WO2023109025A1 (fr)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113918826B (zh) * 2021-12-15 2022-03-25 北京达佳互联信息技术有限公司 投放信息处理方法、资源预测模型训练方法及装置
CN114786031B (zh) * 2022-06-17 2022-10-14 北京达佳互联信息技术有限公司 资源投放方法、装置、设备及存储介质

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110414710A (zh) * 2019-06-20 2019-11-05 平安科技(深圳)有限公司 基于人工智能的趋势预测方法及装置
CN112232854A (zh) * 2020-09-25 2021-01-15 北京三快在线科技有限公司 业务处理方法、装置、设备及存储介质
US20210027379A1 (en) * 2019-07-26 2021-01-28 International Business Machines Corporation Generative network based probabilistic portfolio management
CN112580889A (zh) * 2020-12-25 2021-03-30 北京嘀嘀无限科技发展有限公司 服务资源预估方法、装置、电子设备及存储介质
CN113627979A (zh) * 2021-07-30 2021-11-09 北京达佳互联信息技术有限公司 资源投放数据的处理方法、装置、服务器、系统及介质
CN113918826A (zh) * 2021-12-15 2022-01-11 北京达佳互联信息技术有限公司 投放信息处理方法、资源预测模型训练方法及装置

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11574148B2 (en) * 2018-11-05 2023-02-07 Royal Bank Of Canada System and method for deep reinforcement learning
US11127032B2 (en) * 2018-11-19 2021-09-21 Eventbrite, Inc. Optimizing and predicting campaign attributes
CN112055235B (zh) * 2020-08-25 2022-03-25 北京达佳互联信息技术有限公司 推送展示对象的方法、装置、电子设备及存储介质
CN113570395A (zh) * 2021-01-22 2021-10-29 腾讯科技(深圳)有限公司 信息处理方法、装置、计算机可读介质及电子设备
CN113095885B (zh) * 2021-04-22 2024-04-12 加和(北京)信息科技有限公司 信息投放数据的处理方法和装置
CN113344650B (zh) * 2021-08-05 2021-12-07 北京达佳互联信息技术有限公司 资源数量的确定方法、装置、计算机设备及介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110414710A (zh) * 2019-06-20 2019-11-05 平安科技(深圳)有限公司 基于人工智能的趋势预测方法及装置
US20210027379A1 (en) * 2019-07-26 2021-01-28 International Business Machines Corporation Generative network based probabilistic portfolio management
CN112232854A (zh) * 2020-09-25 2021-01-15 北京三快在线科技有限公司 业务处理方法、装置、设备及存储介质
CN112580889A (zh) * 2020-12-25 2021-03-30 北京嘀嘀无限科技发展有限公司 服务资源预估方法、装置、电子设备及存储介质
CN113627979A (zh) * 2021-07-30 2021-11-09 北京达佳互联信息技术有限公司 资源投放数据的处理方法、装置、服务器、系统及介质
CN113918826A (zh) * 2021-12-15 2022-01-11 北京达佳互联信息技术有限公司 投放信息处理方法、资源预测模型训练方法及装置

Also Published As

Publication number Publication date
CN113918826A (zh) 2022-01-11
CN113918826B (zh) 2022-03-25

Similar Documents

Publication Publication Date Title
WO2023109025A1 (fr) Procédé de traitement d'informations de livraison et procédé et appareil d'entraînement d'un modèle de prédiction de ressources
WO2021135562A1 (fr) Procédé et appareil d'évaluation de validité de caractéristiques, et dispositif électronique et support de stockage
US11288709B2 (en) Training and utilizing multi-phase learning models to provide digital content to client devices in a real-time digital bidding environment
CN108182633B (zh) 贷款数据处理方法、装置、计算机设备和存储介质
JP2022033695A (ja) モデルを生成するための方法、装置、電子機器、記憶媒体、及びコンピュータプログラム製品
CN112905897B (zh) 相似用户确定方法、向量转化模型、装置、介质及设备
CN109389424B (zh) 流量分配方法、装置、电子设备及存储介质
CN110796513A (zh) 多任务学习方法、装置、电子设备及存储介质
CN111124676A (zh) 资源分配方法、装置、可读存储介质和电子设备
CN112015990A (zh) 待推荐网络资源的确定方法、装置、计算机设备及介质
CN113919923B (zh) 直播推荐模型训练方法、直播推荐方法及相关设备
CN113344647B (zh) 一种信息推荐的方法及装置
Jin et al. An intelligent scheduling algorithm for resource management of cloud platform
CN113015010A (zh) 推送参数确定方法、装置、设备及计算机可读存储介质
WO2020211616A1 (fr) Procédé et dispositif de traitement d'informations d'interaction utilisateur
CN113836388A (zh) 信息推荐方法、装置、服务器及存储介质
CN115796937A (zh) 一种大数据复杂关联性的电力供需趋势分析方法和装置
CN115185606A (zh) 业务配置参数的获得方法、装置、设备及存储介质
WO2021240715A1 (fr) Procédé de prédiction d'humeur, dispositif de prédiction d'humeur et programme
CN114529008A (zh) 信息推荐方法、对象识别方法及装置
CN113760550A (zh) 资源分配方法和资源分配装置
CN113010774A (zh) 一种基于动态深度注意力模型的点击率预测方法
CN114764469A (zh) 一种内容推荐方法、装置、计算机设备和存储介质
CN111860870A (zh) 交互行为确定模型的训练方法、装置、设备及介质
CN113365095B (zh) 直播资源推荐方法、装置、电子设备以及存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22905791

Country of ref document: EP

Kind code of ref document: A1