CN-122022927-A - Advertisement data processing method, device, electronic equipment, storage medium and product
Abstract
The invention relates to the technical field of data processing and discloses an advertisement data processing method, an advertisement data processing device, electronic equipment, a storage medium and a product, wherein the method comprises the steps of obtaining historical advertisement bidding data respectively corresponding to each candidate advertisement base price, wherein the historical advertisement bidding data comprises bidding data and overflow price data; and according to the bid probability distribution after fitting and the overflow probability distribution after fitting, evaluating the pricing value of the candidate advertisement reserve price, and determining the target advertisement reserve price from the candidate advertisement reserve prices according to the pricing value of each candidate advertisement reserve price. The invention can improve the configuration efficiency of the advertisement base price and is beneficial to improving the risk of advertisement stream shooting caused by unreasonable advertisement base price configuration.
Inventors
- LI XIAOBO
- LI CHENYU
- ZHANG JIANING
- Mou lei
- BAI XUE
- LUO CHUANJIANG
Assignees
- 杭州网易云音乐科技有限公司
Dates
- Publication Date
- 20260512
- Application Date
- 20251217
Claims (10)
- 1. A method of advertisement data processing, the method comprising: Historical advertisement bidding data corresponding to each candidate advertisement base price respectively are obtained, wherein the historical advertisement bidding data comprise bidding data and overflow data of bidding advertisements with the base price being the candidate advertisement base price for each advertisement demand party; For each candidate advertisement base price, probability distribution fitting is carried out on the bidding probability distribution of the candidate advertisement base price according to the bidding data corresponding to the candidate advertisement base price, and probability distribution fitting is carried out on the overflow probability distribution of the candidate advertisement base price according to the overflow price data corresponding to the candidate advertisement base price; evaluating the pricing value of the candidate advertisement base price according to the fitted bid probability distribution and the fitted overflow probability distribution; And determining target advertisement base prices from the candidate advertisement base prices according to the pricing values of the candidate advertisement base prices.
- 2. The method of claim 1, wherein said probability distribution fitting the bid probability distribution of the candidate advertisement reserve according to the bid data corresponding to the candidate advertisement reserve comprises: and taking whether each advertisement requiring party bids as a random variable of the bidding probability distribution, and carrying out parameter estimation on the bidding probability distribution according to the bidding data corresponding to the candidate advertisement base price.
- 3. The method of claim 2, wherein said parametrically estimating said bid probability distribution from bid data corresponding to said candidate ad reserve comprises: and carrying out parameter estimation on the bid probability distribution through a conjugate prior method according to the bid data corresponding to the candidate advertisement base price.
- 4. The method of claim 1, wherein the probability distribution fitting the overflow probability distribution of the candidate advertisement base according to the overflow data corresponding to the candidate advertisement base comprises: And taking the overflow price of each advertisement requiring party as a random variable of the overflow price probability distribution, and carrying out parameter estimation on the overflow price probability distribution according to the overflow price data corresponding to the candidate advertisement base price.
- 5. The method of claim 4, wherein the parameter estimating the overflow probability distribution according to overflow price data corresponding to the candidate advertisement base price comprises: and carrying out parameter estimation on the overflow probability distribution through a conjugate prior method according to the overflow data corresponding to the candidate advertisement base price.
- 6. The method of claim 1, wherein said evaluating the pricing value of the candidate advertisement base price based on the fitted bid probability distribution and the fitted premium probability distribution comprises: determining a first expected value of the bid probability distribution according to the fitted bid probability distribution; determining a second expected value of the overflow probability distribution according to the fitted overflow probability distribution; and determining the pricing value of the candidate advertisement base price according to the first expected value, the second expected value and the candidate advertisement base price.
- 7. An advertising data processing apparatus, the apparatus comprising: The acquisition module is used for acquiring historical advertisement bidding data corresponding to each candidate advertisement base price respectively, wherein the historical advertisement bidding data comprises bidding data and overflow data of bidding advertisements with the base price being the candidate advertisement base price for each advertisement demand party; The fitting module is used for carrying out probability distribution fitting on the bidding probability distribution of the candidate advertisement base price according to the bidding data corresponding to the candidate advertisement base price and carrying out probability distribution fitting on the overflow probability distribution of the candidate advertisement base price according to the overflow data corresponding to the candidate advertisement base price; the base price evaluation module is used for evaluating the pricing value of the candidate advertisement base price according to the fitted bid probability distribution and the fitted overflow probability distribution; And the base price determining module is used for determining target advertisement base prices from the candidate advertisement base prices according to the pricing values of the candidate advertisement base prices.
- 8. An electronic device, comprising: A memory and a processor in communication with each other, the memory having stored therein computer instructions, the processor executing the computer instructions to perform the advertising data processing method of any one of claims 1 to 6.
- 9. A computer-readable storage medium having stored thereon computer instructions for causing a computer to execute the advertisement data processing method of any one of claims 1 to 6.
- 10. A computer program product comprising computer instructions for causing a computer to perform the advertising data processing method of any one of claims 1 to 6.
Description
Advertisement data processing method, device, electronic equipment, storage medium and product Technical Field The invention relates to the technical field of data processing, in particular to an advertisement data processing method, an advertisement data processing device, electronic equipment, a storage medium and a product. Background In the field of digital advertising, media parties (such as website stations and App developers) can provide advertising services for demand parties by selling advertising spots, bring business growth, and obtain advertising income at the same time so as to win each other. Digital advertisements can be sold in a bid advertisement mode, a demand party needs to bid for each advertisement display opportunity, and a media party carries out advertisement selling according to the principle of' price-high. Typically, to prevent advertisements from being "banked," the media parties will carry an advertisement auction reserve price in each auction bid, thereby telling the requesting parties that the bid must not be lower than the reserve price, i.e., a bid higher than the reserve price, to be able to bid. The reserve price mechanism of advertisement auction, while protecting the media side advertisement from "base sales" to some extent, may also present a risk of streaming beats, e.g., reserve price set too high, without any of the demand side bidding above the reserve price. In the related art, the advertisement reserve price is manually set through manual experience, and in the background of increasing bid advertisement demands, the efficiency of manually configuring the reserve price is low, and the risk of advertisement stream shooting is easily generated due to unreasonable reserve price configuration. Disclosure of Invention The invention provides an advertisement data processing method, an advertisement data processing device, electronic equipment, a storage medium and a product, and aims to improve the risks that advertisement base price configuration efficiency is low and advertisement stream shooting is generated due to unreasonable advertisement base price configuration in the related technology. In a first aspect, the present invention provides an advertisement data processing method, including: Historical advertisement bidding data corresponding to each candidate advertisement base price respectively are obtained, wherein the historical advertisement bidding data comprise bidding data and overflow data of bidding advertisements with the base price being the candidate advertisement base price for each advertisement demand party; For each candidate advertisement base price, probability distribution fitting is carried out on the bidding probability distribution of the candidate advertisement base price according to the bidding data corresponding to the candidate advertisement base price, and probability distribution fitting is carried out on the overflow probability distribution of the candidate advertisement base price according to the overflow price data corresponding to the candidate advertisement base price; evaluating the pricing value of the candidate advertisement base price according to the fitted bid probability distribution and the fitted overflow probability distribution; And determining target advertisement base prices from the candidate advertisement base prices according to the pricing values of the candidate advertisement base prices. In some optional embodiments, the probability distribution fitting the bid probability distribution of the candidate advertisement reserve price according to the bid data corresponding to the candidate advertisement reserve price includes: and taking whether each advertisement requiring party bids as a random variable of the bidding probability distribution, and carrying out parameter estimation on the bidding probability distribution according to the bidding data corresponding to the candidate advertisement base price. In some alternative embodiments, the bid probability distribution is a discrete probability distribution. In some optional embodiments, the estimating the bid probability distribution according to the bid data corresponding to the candidate advertisement base includes: and carrying out parameter estimation on the bid probability distribution through a conjugate prior method according to the bid data corresponding to the candidate advertisement base price. In some optional embodiments, the performing probability distribution fitting on the overflow probability distribution of the candidate advertisement base according to the overflow data corresponding to the candidate advertisement base includes: And taking the overflow price of each advertisement requiring party as a random variable of the overflow price probability distribution, and carrying out parameter estimation on the overflow price probability distribution according to the overflow price data corresponding to the candidate advertisement base price. In so