Click-through rate prediction model integrating user interest and multi-head attention mechanism
Abstract The purpose of click-through rate (CTR) prediction is to anticipate how likely a person is to click on an advertisement or item. It's required for a lot of internet applications, such online advertising and recommendation systems. The previous click-through rate estimation approach suf...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2023-01-01
|
| Series: | Journal of Big Data |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s40537-023-00688-6 |
| _version_ | 1811171676436561920 |
|---|---|
| author | Wei Zhang Yahui Han Baolin Yi Zhaoli Zhang |
| author_facet | Wei Zhang Yahui Han Baolin Yi Zhaoli Zhang |
| author_sort | Wei Zhang |
| collection | DOAJ |
| description | Abstract The purpose of click-through rate (CTR) prediction is to anticipate how likely a person is to click on an advertisement or item. It's required for a lot of internet applications, such online advertising and recommendation systems. The previous click-through rate estimation approach suffered from the following two flaws. On the one hand, input characteristics (such as user id, user age, user age, item id, item category) are usually sparse and multidimensional, making them effective. High-level combination characteristics are used for prediction. Obtaining it manually by domain experts takes a long time and is difficult to finish; also, customer interests are not all the same. The accuracy of the model findings will significantly increase if this immediately recognized component is incorporated in the prediction model. As a consequence, this study creates an IARM (interactive attention rate estimation model) that incorporates user interest as well as a multi-head self-attention mechanism. The deep learning network is used in the model to determine the user's interest expression based on user attributes. The multi-head self-attention mechanism with residual network is then employed to get feature interaction, which enhances the degree of effect of significant characteristics on the estimation result as well as its accuracy. The IARM model outperforms other recent prediction models in the assessment metrics AUC and LOSS, and it has superior accuracy, according to the results from the public experimental data set. |
| first_indexed | 2024-04-10T17:17:56Z |
| format | Article |
| id | doaj.art-0d5b6945ee4845bc99f383409e26d2cd |
| institution | Directory Open Access Journal |
| issn | 2196-1115 |
| language | English |
| last_indexed | 2024-04-10T17:17:56Z |
| publishDate | 2023-01-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | Journal of Big Data |
| spelling | doaj.art-0d5b6945ee4845bc99f383409e26d2cd2023-02-05T12:15:30ZengSpringerOpenJournal of Big Data2196-11152023-01-0110111510.1186/s40537-023-00688-6Click-through rate prediction model integrating user interest and multi-head attention mechanismWei Zhang0Yahui Han1Baolin Yi2Zhaoli Zhang3Faculty of Artificial Intelligence in Education, Central China Normal UniversityFaculty of Artificial Intelligence in Education, Central China Normal UniversityFaculty of Artificial Intelligence in Education, Central China Normal UniversityFaculty of Artificial Intelligence in Education, Central China Normal UniversityAbstract The purpose of click-through rate (CTR) prediction is to anticipate how likely a person is to click on an advertisement or item. It's required for a lot of internet applications, such online advertising and recommendation systems. The previous click-through rate estimation approach suffered from the following two flaws. On the one hand, input characteristics (such as user id, user age, user age, item id, item category) are usually sparse and multidimensional, making them effective. High-level combination characteristics are used for prediction. Obtaining it manually by domain experts takes a long time and is difficult to finish; also, customer interests are not all the same. The accuracy of the model findings will significantly increase if this immediately recognized component is incorporated in the prediction model. As a consequence, this study creates an IARM (interactive attention rate estimation model) that incorporates user interest as well as a multi-head self-attention mechanism. The deep learning network is used in the model to determine the user's interest expression based on user attributes. The multi-head self-attention mechanism with residual network is then employed to get feature interaction, which enhances the degree of effect of significant characteristics on the estimation result as well as its accuracy. The IARM model outperforms other recent prediction models in the assessment metrics AUC and LOSS, and it has superior accuracy, according to the results from the public experimental data set.https://doi.org/10.1186/s40537-023-00688-6User interestMulti-head self-attention mechanismResidual networkClick-through rate prediction model |
| spellingShingle | Wei Zhang Yahui Han Baolin Yi Zhaoli Zhang Click-through rate prediction model integrating user interest and multi-head attention mechanism Journal of Big Data User interest Multi-head self-attention mechanism Residual network Click-through rate prediction model |
| title | Click-through rate prediction model integrating user interest and multi-head attention mechanism |
| title_full | Click-through rate prediction model integrating user interest and multi-head attention mechanism |
| title_fullStr | Click-through rate prediction model integrating user interest and multi-head attention mechanism |
| title_full_unstemmed | Click-through rate prediction model integrating user interest and multi-head attention mechanism |
| title_short | Click-through rate prediction model integrating user interest and multi-head attention mechanism |
| title_sort | click through rate prediction model integrating user interest and multi head attention mechanism |
| topic | User interest Multi-head self-attention mechanism Residual network Click-through rate prediction model |
| url | https://doi.org/10.1186/s40537-023-00688-6 |
| work_keys_str_mv | AT weizhang clickthroughratepredictionmodelintegratinguserinterestandmultiheadattentionmechanism AT yahuihan clickthroughratepredictionmodelintegratinguserinterestandmultiheadattentionmechanism AT baolinyi clickthroughratepredictionmodelintegratinguserinterestandmultiheadattentionmechanism AT zhaolizhang clickthroughratepredictionmodelintegratinguserinterestandmultiheadattentionmechanism |