The widespread adoption of mobile applications has been accompanied by an increasing reliance on thirdparty analytics frameworks for tracking user behavior, monitoring application performance, and enabling monetization. While these frameworks provide essential functionality, they also introduce significant privacy and security risks, particularly when vulnerabilities are present in their implementations. This study analyzes 6,165 APK files using the Mobile Security Framework (MobSF) to identify the top 25 most integrated analytics frameworks, including Google Firebase Analytics, Facebook Login, and Sentry. Leveraging publicly reported Common Vulnerabilities and Exposures (CVEs), this research highlights the security implications associated with these frameworks, focusing on risks such as unauthorized data access, insecure data transmission, and privilege escalation. The analysis reveals systemic vulnerabilities introduced by outdated or improperly implemented frameworks, underscoring the critical need for secure integration practices and timely updates. Furthermore, the findings emphasize the importance of proactive security measures and compliance with privacy regulations to mitigate risks. This study contributes to the understanding of the security landscape in mobile applications, offering insights and recommendations to developers, researchers, and policymakers for enhancing the security and privacy of analytics frameworks.
The rapid proliferation of mobile applications has transformed the way individuals interact with
technology, making apps an integral part of daily life. From social networking and entertainment
to e-commerce and productivity, mobile apps provide a wide array of functionalities. However,
underpinning these user-centric features is a vast ecosystem of third-party analytics frameworks
that collect, analyze, and process user data. These frameworks play a critical role in monitoring
application performance, enhancing user experiences, and enabling monetization through targeted
advertising and personalized content delivery. While analytics frameworks offer substantial
benefits to developers and organizations, they also pose significant privacy and security challenges.
The integration of these third-party frameworks often introduces vulnerabilities into mobile
applications, creating opportunities for unauthorized access to sensitive data, exploitation through
malicious actors, and non-compliance with stringent data protection regulations such as the
General Data Protection Regulation (GDPR) and the California Consumer Privacy Act (CCPA) [
The integration of analytics frameworks within mobile applications has become increasingly
widespread, providing developers with valuable insights into user behavior, application
performance, and overall market trends. These frameworks serve as critical tools for optimizing
user experience and driving business decisions. However, they also introduce complex and
significant concerns regarding data privacy and security. This literature review explores these
issues by synthesizing findings from both academic research and real-world cases, highlighting the
challenges and vulnerabilities posed by data practices within mobile application analytics. Kollnig
et al. (2022) conducted an in-depth examination of Apple’s App Tracking Transparency (ATT) and
privacy labels, mechanisms intended to empower users with greater control over their data. Their
study revealed that while these features have introduced improvements in transparency, many
applications continue to covertly track users, bypassing ATT restrictions through obfuscated
techniques. This finding underscores a critical limitation of platform-level privacy controls,
demonstrating that even robust regulatory interventions can fall short of preventing privacy
violations [
Du et al. (2022) introduced Vizard, an innovative analytics system designed to preserve user privacy by hiding metadata during data analysis. The system enforces fine-grained access control for data owners, addressing the challenge of unauthorized metadata leakage while ensuring efficient analytics [13]. This research underscores the critical need for privacy-by-design solutions in analytics frameworks.
Liu et al. (2023) developed LocationScope, a system for detecting and measuring excessive location data harvesting by mobile apps. Their study revealed how applications often share detailed location data with third-party frameworks, frequently without adequate disclosure or user consent [14]. This work highlights the importance of transparency and accountability in location-based analytics.
Gao et al. (2023) presented ANDetect, a framework for identifying third-party ad networks integrated into Android applications. By analyzing code and network behavior, the framework exposes how ad libraries interact with user data, raising concerns about aggressive data collection practices and the lack of oversight in ad network integration [15].
Another study by Liu et al. (2022) investigated the security risks associated with behavioral data collection by analytics frameworks. Their research identified vulnerabilities in the storage and transmission of user activity data, often exposing sensitive information to exploitation. The authors emphasized the importance of encryption, data minimization, and compliance with privacy regulations [16].
Kumar et al. (2020) explored the vulnerabilities in cloud-based analytics frameworks, which rely on complex interactions between mobile apps and cloud services. They identified insecure APIs, misconfigured permissions, and insufficient access controls as common attack vectors. The study underscores the need for securing both client-side and server-side components of analytics frameworks [17].
Al-Suqri et al. (2017) examined the role of analytics frameworks in mobile commerce applications and their associated security risks. They discussed how improper handling of transaction data by third-party frameworks could result in fraud, data breaches, and privacy violations. The authors proposed practical guidelines for securing analytics frameworks in mobile commerce environments [18].
Taylor et al. (2021) addressed the unique challenges of integrating analytics frameworks into mobile healthcare applications. Their work highlighted vulnerabilities in the handling of sensitive health data, including issues in data transmission, storage, and sharing. They recommended adopting encryption, access controls, and compliance with healthcare-specific regulations such as HIPAA [19].
Dudykevych et al. (2019) conducted a multicriterial analysis to evaluate the efficiency of conservative information security systems, focusing on balancing multiple criteria such as performance, reliability, and resource optimization. Their findings underscore the importance of adopting holistic evaluation frameworks to enhance system security. This approach aligns with the need to assess third-party analytics frameworks comprehensively, considering factors like data privacy, implementation risks, and system integration efficiency [20].
Petrivskyi et al. (2022) introduced a modified methodology for constructing energy-efficient sensor networks by integrating static and dynamic sensors. The optimization of sensor deployment and movement trajectories not only reduced energy consumption but also maintained system effectiveness. These principles can be extended to mobile application ecosystems, where efficient integration and management of analytics frameworks could minimize resource utilization and improve overall performance [21].
Susukailo, Opirsky, and Yaremko (2021) proposed a systematic methodology for establishing Information Security Management Systems (ISMS) to counter modern cybersecurity threats. Their work emphasizes the importance of identifying vulnerabilities, assessing risks, and implementing tailored security controls. These principles are directly applicable to the integration of third-party analytics frameworks, where risk assessment and proactive mitigation strategies are crucial to safeguarding user data [22].
Banakh et al. (2018) explored the detection of MAC spoofing attacks in IEEE 802.11 networks through signal strength analysis. Their methodology demonstrated how inconsistencies in device signal strength could reveal unauthorized activities. Similarly, in mobile application security, anomaly detection techniques could be adapted to identify unusual data flows or unauthorized access attempts within analytics frameworks [23].
Mykhaylova et al. (2023) proposed a method to identify suspicious individuals by analyzing data from mobile devices. Their approach leverages patterns in mobile device usage to detect anomalies that may indicate suspicious behavior. This method underscores the potential of mobile data analytics in enhancing security measures and monitoring individual activities [24].
In a subsequent study, Fedynyshyn, Mykhaylova, and Opirskyy (2024) conducted a static
analysis of Android applications to assess the security implications associated with various mobile
development frameworks. Their findings revealed that certain frameworks might introduce
vulnerabilities, emphasizing the need for developers to carefully consider security aspects when
selecting and implementing these frameworks. The study highlights the importance of thorough
security evaluations in the development process to mitigate potential risks [
These studies collectively highlight the multifaceted privacy and security challenges associated with mobile analytics frameworks. From detecting aggressive data harvesting to addressing vulnerabilities in cloud-based services, the literature emphasizes the importance of secure design, robust regulatory compliance, and innovative privacy-preserving technologies. These findings provide a critical foundation for evaluating the vulnerabilities and risks identified in this study.
This section outlines the methodological framework used to analyze the privacy and security
aspects of analytics packages embedded in mobile applications. The study leverages a dataset of
6,165 APK files subjected to static analysis using MobSF [
This study builds on our previous analysis [
MobSF is a widely used tool for static and dynamic analysis of mobile applications. One of its features is the ability to identify embedded analytical packages within APK files and categorize them based on their functionality and data practices. These categories help classify the different roles of analytics frameworks in mobile applications and their implications for privacy and security. Below, we detail the key categories, their descriptions, and examples.
Crash reporting—as analyzed in the MobSF, crash reporting trackers are components built into
mobile applications to collect and transmit diagnostic data after an application crash. These
trackers are designed to capture critical run-time information, including stack traces, error logs,
device metadata, and user interactions before failure. While such functionality is essential for
developers seeking to increase application stability and optimize performance, it also raises
potential security and privacy concerns. From a cybersecurity perspective, MobSF recognizes that
crash report trackers are a potential data leak point, especially if they collect and transmit sensitive
user information. If improperly configured or inadequately secured, these trackers could expose
personally identifiable information (PII), session details, or other sensitive data to third-party
analytics services. This raises concerns regarding data governance, compliance with privacy
regulations such as GDPR and CCPA, and the risk of unauthorized access by malicious actors. In
addition, integrating third-party crash reporting services such as Firebase Crashlytics [
Profiling—MobSF analysis shows that Profiling trackers exist within mobile applications to
gather and assess user behavior data along with device system parameters and application
operational metrics. The tools serve developers and marketing analysts to investigate application
user habits and detect performance problems while enhancing user interactions. The deployment
of these systems creates potential threats to both data security and user privacy. The MobSF
research profiling trackers a security risk because they gather sensitive data without seeking clear
user authorization. These tracking systems gather information about device location and device
specifications together with processor details and battery status and RAM consumption along with
exclusive device identification numbers. The collection of data without explicit user consent can
result in legal violations of GDPR and CCPA when companies share the information with third
parties for analysis or commercial purposes. When developers integrate third-party profiling
services including Google Analytics [
Advertisement—MobSF examines advertisement trackers which function as components inside mobile applications to track user behavior and gather device information for delivering targeted advertisements. The tracking system collects device identifiers alongside browsing habits and location data and interaction patterns from users before sending this information to advertising networks for relevant ad placement optimization. The essential role advertisement tracking serves for mobile application revenue generation creates major security and privacy risks. Advertisement trackers serve as security risks because they enable unauthorized parties to access and steal user information. Third-party ad networks link up with numerous applications to collect user data for building comprehensive user profiles through their tracking platforms. The absence of proper regulation allows extensive data mining to occur which creates concerns regarding user consent compliance with GDPR and CCPA privacy laws. Some ad trackers obtain persistent device identifiers including IMEI or MAC addresses that enable them to track users across applications even when users adjust their ad personalization controls. Security issues stem from the communication methods that advertisement trackers employ. Due to insecure or unencrypted data transfer methods many of these trackers become vulnerable to interception by attackers. Mobile applications become more vulnerable because of the external advertisement SDKs integrated into their systems. The inclusion of SDKs within applications creates security risks because outdated or insecure code within them exposes the application to threats that include data theft and unauthorized API access and malicious ad network exploitation. MobSF examines network requests and API calls and embedded SDKs to detect advertisement trackers that might endanger user privacy and data security.
Analytics—according to MobSF analytics represents the incorporation of tracking elements into
mobile applications which gather user data throughout interactions to monitor system
performance and analyze user behavior and generate business intelligence. Analytics mechanisms
help developers and organizations understand application usage patterns and discover trends
which allows them to optimize functionality while improving user interaction. From a
cybersecurity standpoint, the extensive data collection performed by analytics services creates
substantial privacy issues, compliance challenges, and security threats. MobSF evaluates analytics
trackers because they function as key channels to collect unauthorized data from users who
provide their sensitive information including location data device identifiers session durations and
in-app interactions. Mobile applications frequently incorporate third-party analytics services
including Google Analytics, Firebase [
Identification—MobSF defines identification as the process of gathering distinct device and user identifiers that mobile applications utilize for authentication procedures and tracking functions and personalized user experiences. Mobile applications track users through both device-specific identifiers such as IMEI and MAC addresses and user-specific information including Android IDs and advertising IDs together with email addresses and phone numbers and account login data. Security identification methods play an important role in protecting users from threats but they pose substantial privacy and security risks when their implementation is not done correctly. MobSF identifies identification tracking as a critical data collection pathway that enables unauthorized profiling of users. A few applications send persistent identifiers to external services which allows third parties to track users through multiple programs. Multiple device parameters combined through this practice generate unique user profiles that occur without user consent. The implementation of tracking systems violates user privacy and violates GDPR and CCPA because these regulations specify detailed rules about user data collection and storage. The storage or transmission of identifiers by applications generates security problems because of insufficient encryption and weak access control measures. The exposure of identifiers through unsecured network requests and weak local storage allows criminals to intercept and manipulate and misuse these identifiers. The exposure of identifiers enables attackers to commit identity theft and unauthorized account entry and perform device impersonation attacks. MobSF examines application identification methods through API call analysis and data storage tracking and network protocol examination to determine security and privacy risks.
Location—MobSF defines location as the process of collecting and processing and transmitting geospatial data from mobile devices to identify user location. Mobile applications collect location data using GPS and Wi-Fi signals and cell tower information and Bluetooth to deliver services like navigation, targeted advertising, geofencing, and content personalization. The necessity of location data for improving user experience generates serious privacy and security challenges that mobile application security must actively handle. According to MobSF location tracking stands as a fundamental security risk since numerous applications gather user-sensitive information without proper consent or notification. A significant number of mobile applications that use analytics and advertising features collect location data without notifying users directly about this information collection process. The collected data enables the creation of detailed user profiles through geolocation patterns but presents privacy risks when disclosure and management are inadequate. Location data becomes a target for malicious actors because they use it to monitor people and identify valuable targets and execute social engineering schemes. Unprotected channels for transmitting location data expose users to interception risks, especially during transmissions of sensitive geospatial information to unencrypted third-party servers. Many applications that request access to user locations do not implement adequate security measures to protect this data from unauthorized exposure during storage or transmission. The location data review at MobSF involves checking location permission levels and network traffic analysis for unauthorized signals and encryption and data management practice assessments to evaluate security and privacy risks.
The analysis of 6,165 APK files scanned with MobSF reveals comprehensive insights into the usage patterns, categories, and implications of analytics frameworks and trackers embedded in mobile applications. By examining individual tracker popularity, functional categories, and app genrespecific trends, this section provides a detailed understanding of the privacy and security challenges associated with these frameworks.
The analysis identified the most frequently used trackers in the scanned APK files. Among the findings:
Google Firebase Analytics emerged as the most commonly integrated framework, appearing in 5,207 apps (84.46%). Its widespread usage reflects its dual role as a tool for user behavior analysis and app performance monitoring.
Google CrashLytics was present in 3,687 apps (59.81%), underlining its critical function in debugging and crash reporting.
Google AdMob [
Facebook Login [
Facebook Share [
These findings emphasize the dominance of Google and Facebook frameworks across various functionalities, raising concerns about data centralization and potential privacy risks. Smaller trackers, while less prevalent, also contribute to the ecosystem, with varying implications based on their functionality.
The list of the top 25 most integrated analytical frameworks is provided in Table 1.
Trackers were categorized based on their primary roles in mobile applications (see Fig. 1). The analysis revealed the following key categories:
Analytics—representing the largest category, 127 unique trackers were identified. These frameworks provide insights into user behavior, app performance, and engagement metrics. They are indispensable for developers but pose risks of excessive data collection. Advertisement—with 77 unique trackers, this category underscores the focus on monetization through ad targeting. These trackers often collect detailed user profiles, including location and behavioral data.
Profiling—comprising 33 unique trackers, these frameworks specialize in gathering user attributes for personalized experiences and targeted advertising. This category raises significant privacy concerns due to the detailed nature of the data collected.
Crash Reporting—with 15 unique trackers, these frameworks support app stability by logging errors and crashes. While useful for debugging, they may inadvertently capture sensitive data in logs if not configured properly.
Location Services—consisting of 25 unique trackers, these frameworks enable geolocationbased functionalities. Apps using these trackers must carefully balance functionality with privacy to avoid exposing sensitive user data.
These categories illustrate the diversity of data collection practices in mobile apps and highlight specific privacy and security risks associated with particular functionalities.
The prevalence and average number of trackers were analyzed across different app genres, revealing notable trends:
Dating Apps—with an average of 7.64 trackers per app across 202 samples, dating apps exhibited the highest reliance on analytics and advertising frameworks. This trend reflects the genre’s focus on engagement and monetization, but it also raises significant concerns about the handling of sensitive user data.
Personalization Apps—these apps had an average of 7.11 trackers per app (274 samples), indicating extensive use of profiling and analytics to tailor user experiences.
News and Magazines—with 7.05 trackers per app (443 samples), this genre heavily integrates advertising frameworks to support revenue generation through targeted ads. Entertainment Apps—averaging 6.60 trackers per app (417 samples), these apps focus on engagement metrics, often integrating social media and advertising frameworks. Social Media Apps—with 6.35 trackers per app (438 samples), this genre relies on analytics, profiling, and location services to optimize user interactions and ad targeting.
See extended data on the average number of trackers per Google Play genre in Fig. 2. This variability across genres reflects differing priorities in app development, with some genres placing a greater emphasis on data collection to support monetization and user engagement.
The dominance of specific companies and frameworks in the tracker ecosystem emerged as a key theme. Google and Facebook SDKs accounted for a substantial portion of all identified trackers, providing critical functionality in analytics, advertising, and social media integration. However, this concentration of power also creates systemic risks, including:
Centralization of user data—the reliance on a few dominant players amplifies risks of extensive data aggregation, which could be exploited in case of data breaches or unauthorized access.
Privacy risks in profiling and location services—trackers in the profiling and location services categories were observed to gather highly sensitive user information. These practices raise ethical and legal concerns, particularly in light of regulations like GDPR and CCPA.
Potential security vulnerabilities—frameworks with known vulnerabilities, such as outdated SDKs, increase the attack surface for mobile applications. Apps that fail to update their trackers to patch known vulnerabilities expose themselves and their users to potential exploitation.
The static analysis of 6,165 APK files using MobSF revealed the top 25 most integrated analytics frameworks in mobile applications. While these frameworks provide essential functionalities such as user behavior tracking, crash reporting, and monetization, their widespread adoption introduces significant privacy and security risks. This discussion focuses on the vulnerabilities (CVEs) identified for these frameworks, highlighting their implications for developers, users, and application ecosystems.
Google’s analytics and advertising frameworks dominate mobile app ecosystems, with Firebase Analytics (84.46%), CrashLytics (59.81%), and AdMob (50.07%) being the most integrated frameworks in the dataset. Although these tools are indispensable for monitoring performance and generating revenue, their extensive use raises critical security concerns:
Google Firebase Analytics has been associated with improper implementation risks. While no specific CVEs have been reported, insecure configurations or insufficient access controls can result in unauthorized data exposure.
Google CrashLytics handles crash logs that may inadvertently include sensitive user information. Although no direct CVEs exist, improper sanitization or logging practices can expose apps to significant data leakage risks.
Google AdMob facilitates in-app advertisements, but its integration increases the risk of user profiling and exposure to malicious ads. Historical vulnerabilities in Google’s advertising ecosystem, such as exposed APIs, have demonstrated the need for secure implementation and regular monitoring.
These findings underscore the need for rigorous data minimization practices and periodic security audits for applications integrating Google frameworks.
Facebook’s SDKs are among the most widely adopted, with Facebook Login (32.38%), Share (29.59%), and Analytics (26.65%) leading the list. However, these frameworks have historically been targets for security exploits:
Facebook Login—while offering seamless user authentication, vulnerabilities such as improper session handling or token hijacking have been reported in similar authentication systems. For example, CVE-2018-6013 highlights the risks of session misuse.
Facebook Analytics and Ads—these frameworks collect detailed user data for engagement and advertising purposes. Although no recent CVEs are directly linked to Facebook Ads, concerns about privacy breaches through excessive data collection persist, particularly given the platform’s past involvement in large-scale data misuse cases (e.g., Cambridge Analytica).
Facebook Share—while no specific vulnerabilities have been disclosed, insecure implementations could expose user data shared via social media, leading to privacy violations.
The significant reliance on Facebook frameworks underscores the importance of ensuring secure SDK versions and adhering to privacy-by-design principles.
Sentry (5.27%) [
CVE-2025-22146 [
CVE-2024-45606 [
CVE-2023-46729 [
These vulnerabilities highlight the risks of relying on third-party monitoring tools for critical operations. Ensuring timely patching and secure integration is essential to mitigating such risks.
AppsFlyer [
IAB Open Measurement [
Frameworks such as Unity3d Ads [
The integration of third-party analytics frameworks introduces systemic risks to mobile applications:
Privacy Risks—frameworks that collect extensive user data, particularly for profiling and advertising, can lead to unauthorized data sharing and non-compliance with regulations like GDPR and CCPA.
Increased Attack Surface—each integrated framework introduces potential vulnerabilities. Applications with multiple frameworks amplify this risk, particularly if outdated SDK versions are used.
Regulatory Challenges—misaligned data collection practices and insufficient transparency can expose developers to legal and reputational risks.
These results provide a comprehensive understanding of the prevalence and impact of trackers in mobile applications. By analyzing their popularity, functional categorization, and usage across genres, this study reveals critical insights into the privacy and security challenges posed by analytics frameworks, paving the way for further discussions on mitigation strategies and regulatory improvements.
Declaration on Generative AI While preparing this work, the authors used the AI programs Grammarly Pro to correct text grammar and Strike Plagiarism to search for possible plagiarism. After using this tool, the authors reviewed and edited the content as needed and took full responsibility for the publication’s content. [6] X. Liu, et al., Privacy risk analysis and mitigation of analytics libraries in the Android ecosystem, in: IEEE Transactions on Mobile Computing, vol. 19(5), 2020, 1184–1199. doi:10.1109/TMC.2019.2903186 [7] A. Nyzhnyk, A. Partyka, M. Podpora, Increase the cybersecurity of SCADA and IIoT devices with secure memory management, in: Cyber Security and Data Protection, vol. 3800, 2024, 32– 41. [8] Y. Martseniuk, et al., Shadow IT risk analysis in public cloud infrastructure, in Cyber Security and Data Protection, vol. 3800, 2024, 22–31. [9] Facebook to pay $725 million to settle lawsuit over Cambridge Analytica data leak. URL: https://thehackernews.com/2022/12/facebook-to-pay-725-million-to-settle.html [10] 540 million facebook user records found on unprotected Amazon servers. URL: https://thehackernews.com/2019/04/facebook-app-database.html [11] More than 250 iOS Apps caught using private APIs to collect users’ private data. URL: https://thehackernews.com/2015/10/apple-ios-malware-apps.html [12] Warning! Popular Apple Store Apps infected with data-theft malware. URL: https://thehackernews.com/2015/09/apple-apps-malware.html [13] C. Cai, et al., Vizard: A metadata-hiding data analytic system with end-to-end policy controls, in: 2022 ACM SIGSAC Conference on Computer and Communications Security (CCS’22), 2022, 441–454. doi:10.1145/3548606.3559349 [14] H. Lu, et al., Detecting and measuring aggressive location harvesting in mobile apps via dataflow path embedding, in: Proc. ACM Meas. Anal. Comput. Syst. 7(1), 2023. doi:10.1145/3579447 [15] L. Xinyu, et al., ANDetect: A third-party ad network libraries detection framework for Android applications, in: 39th Annual Computer Security Applications Conference (ACSAC ‘23), 2023, 98–112. doi:10.1145/3627106.3627182 [16] Y. Han, et al., Systematic analysis of security and vulnerabilities in miniapps, in: 2023 ACM Workshop on Secure and Trustworthy Superapps (SaTS’23), 2023, 1–9. doi:10.1145/3605762.3624432 [17] S. Canard, et al., WeStat: A privacy-preserving mobile data usage statistics system, in: 2021 ACM Workshop on Security and Privacy Analytics (IWSPA’21), 2021, 5–14. doi:10.1145/3445970.3451151 [18] I. Kuksa, et al., Chapter five—Predictive personalisation: are we watching or being watched?,
Understanding Personalisation (2023) 89–108. doi:10.1016/B978-0-08-101987-0.00008-4 [19] L. K. Agrawal, D. K. Agrawal, K. G. Srinivasa, Security and privacy in health data storage and its analytics, Blockchain for 5G Healthcare Applications: Security and privacy solutions, (2024) 249–285. doi:10.1049/PBHE035E_ch10 [20] V. Dudykevych, et al., A multicriterial analysis of the efficiency of conservative information security systems, Eastern-European J. Enterp. Technol. (2019) 6–13. doi:10.15587/17294061.2019.166349 [21] V. Petrivskyi, et al., Development of a modification of the method for constructing energyefficient sensor networks using static and dynamic sensors, Eastern-European J. Enterp.
Technol. 1.9(115) (2022) 15–23. doi:10.15587/1729-4061.2022.252988 [22] V. Susukailo, I. Opirsky, O. Yaremko, Methodology of ISMS establishment against modern cybersecurity threats, in: Future Intent-Based Networking. Lecture Notes in Electrical Engineering, vol. 831, 2021, 257–271. doi:10.1007/978-3-030-92435-5_15 [23] R. Banakh, A. Piskozub, I. Opirskyy, Detection of MAC spoofing attacks in IEEE 802.11 networks using signal strength from attackers’ devices, in: Advances in Computer Science for Engineering and Education, ICCSEEA 2018, Advances in Intelligent Systems and Computing, vol. 754, 2018, 468–477. doi:10.1007/978-3-319-91008-6_47 [24] T. Fedynyshyn, I. Opirskyy, O. Mykhaylova, A method to detect suspicious individuals through mobile device data, in: 2023 IEEE 5th International Conference on Advanced Information and Communication Technologies (AICT), 2023, 82–86. doi:10.1109/AICT61584.2023.10452683