Source-linked AI summary
Beyond Google Play: A Large-Scale Comparative Study of Chinese Android App Markets
Haoyu Wang, Zhe Liu, Jingyue Liang, Narseo Vallina-Rodriguez, Yao Guo, Li Li, Juan Tapiador, Jingcun Cao, Guoai Xu
TL;DR
The paper addresses the limited comparative evidence on China’s fragmented Android app-market ecosystem. It analyzes millions of apps across Chinese markets and Google Play, finding higher prevalence of malicious and deceptive apps in Chinese stores and substantial market differences.
Problem
The Chinese Android ecosystem has received limited large-scale comparative study, despite its diverse third-party markets and risks from malicious and repackaged apps.
Method
The study compares over 6 million Android apps from 16 Chinese markets and Google Play across market features, publishing dynamics, developer behavior, and app misbehavior.
Results
Chinese markets show higher prevalence of fake, cloned, and malicious apps than Google Play, while market behaviors and third-party library use differ substantially.
Takeaways & Limitations
The findings motivate greater research, user and developer awareness, regulatory attention, and better operational practices among app-store operators.
Takeaways & Limitations
Third-party-library analysis relies on detection resources that were too old or incomplete for the study’s purpose, including a library dataset derived from Google Play.
Abstract
from arXiv · showhide
China is one of the largest Android markets in the world. As Chinese users cannot access Google Play to buy and install Android apps, a number of independent app stores have emerged and compete in the Chinese app market. Some of the Chinese app stores are pre-installed vendor-specific app markets (e.g., Huawei, Xiaomi and OPPO), whereas others are maintained by large tech companies (e.g., Baidu, Qihoo 360 and Tencent). The nature of these app stores and the content available through them vary greatly, including their trustworthiness and security guarantees. As of today, the research community has not studied the Chinese Android ecosystem in depth. To fill this gap, we present the first large-scale comparative study that covers more than 6 million Android apps downloaded from 16 Chinese app markets and Google Play. We focus our study on catalog similarity across app stores, their features, publishing dynamics, and the prevalence of various forms of misbehavior (including the presence of fake, cloned and malicious apps). Our findings also suggest heterogeneous developer behavior across app stores, in terms of code maintenance, use of third-party services, and so forth. Overall, Chinese app markets perform substantially worse when taking active measures to protect mobile users and legit developers from deceptive and abusive actors, showing a significantly higher prevalence of malware, fake, and cloned apps than Google Play.
1 INTRODUCTION
Chinese Android users rely on many alternative app markets because Google Play is inaccessible, creating a fragmented ecosystem that is difficult to navigate and vulnerable to malicious distribution. This study compares these markets at scale and across multiple dimensions.
- 1 INTRODUCTION: More than 700 million Android users in China rely on alternative markets because Google services, including Google Play, are restricted.These markets are operated by both Chinese technology companies and smartphone manufacturers.
- 1 INTRODUCTION: The diversity of third-party markets makes it difficult for users and developers to choose where to discover or distribute apps.The fragmented ecosystem also creates opportunities for repackaged apps and malware to spread.
- 1 INTRODUCTION: The study compares Chinese app markets with Google Play across intra- and inter-market features, publishing behavior, and app misbehavior.It examines more than 6 million apps and reports heterogeneous market and developer behaviors.
- 1 INTRODUCTION: The study is presented as the first large-scale, longitudinal comparison of Google Play and alternative Chinese app stores across multiple dimensions.The authors released their dataset and experimental results to the research community.
2 CHINESE ANDROID APP MARKETS
Chinese Android markets include vendor-specific stores, web-company stores, and specialized markets, with differing access policies, incentives, auditing practices, and transparency requirements. Compared with Google Play, their governance mechanisms vary substantially.
- 2 CHINESE ANDROID APP MARKETS: Chinese app markets comprise vendor-specific stores, web-company stores, and specialized markets, with the study covering the top ten Chinese markets except Vivo.The sample includes stores operated by major smartphone vendors and Chinese web companies.
- 2.1 Features of Chinese App Markets: Most Chinese markets let third-party developers publish apps, but some restrict eligibility, app categories, device compatibility, or APK size.Lenovo limits publishing to registered companies, OPPO limits categories, vendor stores enforce device compatibility, and App China caps APKs at 50 MB.
- 2.1 Features of Chinese App Markets: Chinese stores use publication incentives such as temporary exclusivity and free promotion to attract developers and encourage app publication.These mechanisms include starting-app and exclusive-app promotion and high-quality-app promotion.
- 2.1 Features of Chinese App Markets: All Chinese markets except HiApk and PC Online perform copyright ownership checks, while most markets claim to inspect apps before publication.Eight markets claim to combine automated security analysis with human inspection.
- 2.1 Features of Chinese App Markets: None of the Chinese markets require privacy policies for apps handling sensitive user data, although nine report advertisements and only Google Play and 360 report in-app purchases.This creates a marked transparency difference between Chinese markets and Google Play.
3 APK COLLECTION
The study collected APKs and public metadata from Google Play and 16 Chinese Android markets using market-specific crawlers and parallel searching. The resulting dataset emphasizes popular apps while spanning millions of apps and APK files.
- 3 APK COLLECTION: For each app, the study collected public metadata including names, versions, categories, descriptions, downloads, ratings, and release or update dates.APK harvesting covered Google Play and the 16 alternative Chinese stores listed in the study.
- 3 APK COLLECTION: The researchers adapted crawling strategies to each market’s indexing behavior and used parallel searches to identify matching apps across stores.The crawl ran in parallel across 50 Aliyun Cloud Servers from August 15 to August 30, 2017.
- 3 APK COLLECTION: The crawlers collected metadata for 6,267,247 apps and 4,522,411 APK files across Google Play and 16 Chinese markets.The dataset is described as the largest cross-store APK collection obtained by the research community.
- 3 APK COLLECTION: The dataset does not cover every available app because of BFS crawling limits, but the authors believe it covers the most popular apps and avoids cross-market bias through parallel searching.Google Play APK gaps were partly addressed using AndroZoo, while rate limiting restricted direct APK collection.
4 GENERAL OVERVIEW
Across Google Play and 16 Chinese markets, app catalogs show substantial differences in popularity, platform support, maintenance, third-party services, and user ratings. Chinese markets collectively have major reach but generally show older apps, weaker metadata quality, and less informative user feedback.
- 4.1 App Categories: Games comprise roughly 50% of apps across markets, although vendor stores such as Meizu, Huawei, and Lenovo show different category distributions.
- 4.2 User Downloads: 193 billion downloads were aggregated in Google Play, while the 16 Chinese markets together reached three times that volume.
- 4.2 User Downloads: The top 0.1% of apps account for more than 50% of downloads in every market, exceeding 80% in Tencent MyApp.
- 4.3 Minimum API Level: Chinese alternative stores support low Android API levels at three times the rate of Google Play, consistent with their older release and update profiles.
- 4.3 Minimum API Level: Only 5% of Chinese-store apps were released or updated within six months of crawling, compared with more than 23% on Google Play.
- 4.4 Third-party Libraries: Third-party libraries appear in roughly 94% of Google Play apps and 85% of PC Online apps, while Chinese markets also commonly embed WeChat, Alipay, and Baidu libraries.
- 4.4 Third-party Libraries: More than 80% of apps in several Chinese markets have no user ratings, whereas only 9.3% of Google Play apps lack ratings and more than 50% receive scores above 4.
5 PUBLISHING DYNAMICS
Developers target Chinese and global app stores unevenly, producing heterogeneous publishing, version-maintenance, and catalog-sharing patterns across markets.
- Measurement: The study identifies slightly over 1 million app developers using distinct APK developer signatures.Developer signatures were extracted with ApkSigner; package names identify unique apps across markets.
- Developer publishing strategies: 57% of developers found on Google Play do not release their apps in Chinese markets, while around 48% focus solely on Chinese alternatives.More than half release on Google Play, revealing limited overlap between ecosystems.
- Developer publishing strategies: Around 20% of developers publish in more than three stores, but only 696 distribute apps across all 17 markets.Over 10% target exactly one Chinese store, especially in larger catalogs such as Tencent and 25PP.
- Single-store apps: 77% of Google Play apps are single-store, compared with an 11% average across Chinese alternative stores, varying from over 20% to under 1%.AnZhi, OPPO, and 25PP exceed 20%, whereas Wandoujia and Meizu remain below 1%.
- Catalog overlap: Over 80% of the top 1% most popular apps are shared across Chinese markets, but similarity between Chinese stores and Google Play is low.This indicates strong sharing among Chinese stores alongside substantial ecosystem separation from Google Play.
- Version maintenance: Roughly 14% of apps publish multiple versions simultaneously across different stores, with extreme cases reaching 14 versions.The analysis relaxes version matching because unsynchronized roll-outs can otherwise obscure multi-store releases.
6 DEVELOPER MISBEHAVIORS
The study measures fake, cloned, over-privileged, and malware-related behavior using clustering, signature and code analysis, permission analysis, and antivirus aggregation.
- Fake apps: Fake-app detection clusters identical names across different package names, but legitimate multi-platform or generic-name apps require heuristic removal.Manual analysis found 83% of fake apps in small clusters containing unpopular apps and a popular official app.
- Cloned apps: 12% of apps have at least two signature-based clones, and manual examination found all 100 sampled pairs were repackaged apps.The app com.dino.dinosuperapp appeared in 15 markets under 11 developer signatures.
- Cloned apps: Code-based clones affect roughly 20% of apps versus roughly 10% for signature-based clones, with Google Play the premier source for cloning into Chinese markets.LibRadar reduces third-party-library effects, while Figure 10 maps intra- and inter-market clone destinations.
- Over-privileged apps: Chinese markets generally contain more over-privileged apps than Google Play, where approximately 65% of apps are over-privileged.The analysis groups apps by excess sensitive permissions labeled dangerous by Google.
- Malware prevalence: For AV-rank ≥10, malware exceeds 10% in 11 of 16 Chinese markets, reaching more than 24% in PC Online versus 2.09% in Google Play.Huawei’s market is comparatively lower at 4.71%, while Tencent and 25PP host the largest absolute counts.
7 POST-ANALYSIS
A second crawl evaluated whether markets removed potentially malicious apps, revealing strong removal in Google Play but limited and uneven removal across Chinese stores.
- Malware removal: Over 84% of potentially malicious apps found in Google Play were removed, while Chinese-market malware removal ranged from 0.01% to 34.51%.HiAPk was excluded because it discontinued service, and OPPO was inaccessible except through its market app.
- Malware families: Figure 12 compares the distribution of the top 15 malware families between Google Play and Chinese markets.The supplied caption establishes the comparison scope but not the outcome.
- Cross-market persistence: More than 70% of 11,623 apps removed from Google Play were still hosted by at least one Chinese market by April 2018.Tencent and PC Online were the Chinese stores where these potentially malicious apps most clearly persisted.
8 DISCUSSION
Chinese app markets show more frequent malicious and deceptive activity than Google Play, but Huawei and Lenovo exhibit stricter security behavior while retaining outdated apps.
- 8 DISCUSSION: Chinese markets show more common potentially malicious and deceptive activity than Google Play, with Tencent Myapp and PC Online particularly indulgent toward abusive apps.Tencent Myapp hosts substantial malware, while PC Online shows similar behavior without an identified pre-publication security policy.
- 8 DISCUSSION: Figure 13 normalizes each market metric to a 0–100 scale around a zero center for multidimensional comparison.The comparison covers Google Play, Tencent, PC Online, Huawei, and Lenovo markets.
- 8 DISCUSSION: Huawei and Lenovo label 4.71% and 7.53% of their apps as malware, respectively, comparable in magnitude to Google Play.Both stores publish popular apps and show similar ratings and download distributions, alongside stricter mechanisms against malware distribution.
- 8 DISCUSSION: Despite stronger malware controls, Huawei and Lenovo retain many outdated apps, potentially limiting access to new features and bug fixes.Lenovo’s restriction of publishing to registered companies may help mitigate malware and low-quality apps, while outdated software can hurt perceived quality and brand equity.
9 RELATED WORK
Prior research has extensively analyzed mobile apps and Google Play, but Chinese alternative Android markets remain comparatively understudied. This work addresses that gap through a large-scale comparative analysis of Google Play and Chinese markets.
- 9 RELATED WORK: The study is presented as the largest and most exhaustive comparison between Google Play and Chinese alternative markets.The authors position the work as addressing an ecosystem that previous research had not studied well.
- 9 RELATED WORK: AndroZoo compiled more than 5M APKs from 12 app stores, including five Chinese markets, enabling studies of malicious practices and privacy risks.Its Chinese-market subset contains roughly 1.5 million apps.
- 9 RELATED WORK: Google Play research has examined large-scale app characterization, temporal patterns, app evolution, authentication schemes, and market attributes.Examples include PlayDrone’s analysis of 1.1 million apps and a six-month daily analysis of 160,000 apps.
10 CONCLUSION
The paper analyzes over 6 million apps from 16 Chinese markets and Google Play to compare the Chinese Android ecosystem with Google Play. It finds substantial ecosystem differences alongside some minor commonalities.
- 10 CONCLUSION: The analysis covers over 6 million Android apps obtained from 16 Chinese app markets and Google Play.The study examines app-store features and ecosystem differences at large scale.
- 10 CONCLUSION: The Chinese Android ecosystem differs substantially from Google Play, although the study also identifies some minor commonalities.