KLAN: Kuaishou Landing-page Adaptive Navigator

TL;DR

KLAN (Kuaishou Landing-page Adaptive Navigator), a hierarchical solution framework designed to provide personalized landing pages under the formulation of PLPM, is proposed, a hierarchical solution framework designed to provide personalized landing pages under the formulation of PLPM.

Abstract

Modern online platforms configure multiple pages to accommodate diverse user needs. This multi-page architecture inherently establishes a two-stage interaction paradigm between the user and the platform: (1) Stage I: page navigation, navigating users to a specific page and (2) Stage II: in-page interaction, where users engage with customized content within the specific page. While the majority of research has been focusing on the sequential recommendation task that improves users' feedback in Stage II, there has been little investigation on how to achieve better page navigation in Stage I. To fill this gap, we formally define the task of Personalized Landing Page Modeling (PLPM) into the field of recommender systems: Given a user upon app entry, the goal of PLPM is to proactively select the most suitable landing page from a set of candidates (e.g., functional tabs, content channels, or aggregation pages) to optimize the short-term PDR metric and the long-term user engagement and satisfaction metrics, while adhering to industrial constraints. Additionally, we propose KLAN (Kuaishou Landing-page Adaptive Navigator), a hierarchical solution framework designed to provide personalized landing pages under the formulation of PLPM. KLAN comprises three key components: (1) KLAN-ISP captures inter-day static page preference; (2) KLAN-IIT captures intra-day dynamic interest transitions and (3) KLAN-AM adaptively integrates both components for optimal navigation decisions. Extensive online experiments conducted on the Kuaishou platform demonstrate the effectiveness of KLAN, obtaining +0.205% and +0.192% improvements on in Daily Active Users (DAU) and user Lifetime (LT). Our KLAN is ultimately deployed on the online platform at full traffic, serving hundreds of millions of users. To promote further research in this important area, we will release our dataset and code upon paper acceptance.

Figures (9)

• Figure 1: An example of the two-stage interaction paradigm on multi-page platforms: the user is first navigated to a functional or channel page, and then interacts with content within that page.
• Figure 2: (a) Users' long-term metric performance across different active page counts, where core users are those whose time spent on a page exceeds a certain percentile. Full attendance ratio denotes the proportion of the users who are active each day within the statistical period and LT is an unbiased estimate of medium and long-term DAU for experimental evaluation (as detailed in Equation \ref{['equ:LT7']}). Here we use 7-day Lifetime and LTR is calculated with a normalization operation ($\frac{\textbf{LT}}{\textbf{7}}$); (b) The proportion of multi/single-page users over 14 consecutive days; (c) The distribution of users based on their number of app entry within a single day.
• Figure 3: The architecture of our proposed method.
• Figure 4: Illustration of the online pipeline.
• Figure 5: Online A/B test experimental results of DAU and LT.