[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-126223-en":3,"doc-seo-126223-105":30,"detail-sidebar-cat-0-en-105":92},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":11,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":27,"seo_description":14,"update_tm":28,"read_time":29},126223,549768702563,"Sage","https://ap-avatar.wpscdn.com/avatar/8000c4aa63b76e948b?x-image-process=image/resize,m_fixed,w_180,h_180&k=1786536092046926083",8,"Research & Report","Explainable Machine Learning for Performance Anomaly Detection and Classification in Mobile Networks","Mobile communication providers collect extensive parameters, statistics, and KPIs to identify operating scenarios that degrade Internet-based service quality. Anomaly detection and classification in mobile networks are difficult because of the large number of variables and the unknown feature distributions. The work proposes XMLAD, an unsupervised methodology combining a data-cleaning stage with explainable machine learning models. It removes outliers to train an anomaly detector and then builds an anomaly classification engine using discretized KPI differences and prediction labels. Two decision tree classifiers produce interpretable feature and threshold explanations for normal and anomalous behaviors. Evaluation uses real commercial-network datasets and a synthetic testbed based on a known TCP throughput model to validate accuracy.","Explainable Machine Learning for Performance Anomaly Detection and  \nClassi􀀌cation in Mobile Networks ⋆  \nJuan M. Ram􀀓􀀐reza,∗, Fernando D􀀓􀀐ezb , Pablo Rojoc , Vincenzo Mancusoa , Antonio Fern􀀓andez-Antaa  \naIMDEA Networks Institute, 28918, Madrid,  \nb Universidad Politecnica de Madrid,  \ncNokia CNS,  \nAbstract  \nMobile communication providers continuously collect many parameters, statistics, and key performance indicators (KPIs) with the goal of identifying operation scenarios that can a􀀋ect the quality of Internet-based services. In this regard, anomaly detection and classi􀀌cation in mobile networks have become challenging tasks due to both the huge number of involved variables and the unknown distributions exhibited by input features. This paper introduces an unsupervised methodology based on both a data-cleaning strategy and explainable machine learning models to detect and classify performance anomalies in mobile networks. Speci􀀌cally, this methodology dubbed explainable machine learning for anomaly detection and classi􀀌cation (XMLAD) aims at identifying features and operation scenarios characterizing performance anomalies without resorting to parameter tuning. To this end, this approach includes a data cleaning stage that extracts and removes outliers from experiments and features to train the anomaly detection engine with the cleanest possible dataset. Moreover, the methodology considers the di􀀋erences between discretized values of the target KPI and labels predicted by the anomaly detection engine to build the anomaly classi􀀌cation engine which identi􀀌es features and thresholds that could cause performance anomalies. The proposed methodology incorporates two decision tree classi􀀌ers to build explainable models of anomaly detection and classi􀀌cation engines whose decision structures recognize features and thresholds describing both normal behaviors and performance anomalies. We evaluate the XMLAD methodology on real datasets captured by operational tests in commercial networks. In addition, we present a testbed that generates synthetic data using a known TCP throughput model to assess the accuracy of the proposed approach.  \nKeywords: anomaly detection and classi􀀌cation, data cleaning, decision tree classi􀀌ers, explainable machine learning, mobile networks.  \n1. Introduction  \nMobile networks have shown remarkable growth over the last two decades. More precisely, 􀀌fth-generation (5G) mobile technologies have played an important role in wireless network access providing high-speed connectivity for an increasing number of heterogeneous devices [2, 3] . With the consolidation of the 5G and the rising of the 6G, mobile networks exhibit increasingly complex structures, with distributed servers operating in a coordinated way to ensure reliable network services. With mobile networks becoming increasingly complex, it has become necessary to develop di􀀋erent approaches based on learning architectures to solve operating problems [4], such as network tra􀀎c [5], dynamic spectrum access [6] and energy-e􀀎cient computation [7] . In this context, mobile network providers continuously collect information about the communication system  \n⋆ A preliminar version of this manuscript is included in the proceedingns of the 20th Mediterranean Communication and Computer Networking Conference (MedComNet 2022) [1] .  \n∗ Corresponding author.  \n[Email address:](Email address: juan.ramirez@imdea.org)[ juan.ramirez@imdea.org](Email address: juan.ramirez@imdea.org) (Juan M. Ram􀀓􀀐rez) Preprint submitted to Computer Communications  \nwith the aim of detecting operation instances that could unveil the misfunctioning of di􀀋erent network components. However, network health diagnosis and anomaly identi􀀌cation have become challenging tasks because of the increasing structural complexity of mobile networks requiring a large number of variables to monitor the communication system performance.  \nIn the context of computer networks, anomalies can be categorized in","cbCaicIX4DyPxKJK","https://ap.wps.com/l/cbCaicIX4DyPxKJK","pdf",13260600,1,22,"English","en",105,"# Introduction\n## Motivation for mobile network anomaly detection\n## Related work and limitations\n# Proposed XMLAD methodology\n## Data cleaning and outlier removal\n## Anomaly detection engine\n## Anomaly classification engine\n## Explainable decision-tree models\n# Evaluation\n## Experiments on operational datasets\n## Synthetic data testbed using TCP throughput model","[{\"question\":\"Why are anomaly detection and classification challenging in mobile networks?\",\"answer\":\"They require handling many monitored variables and coping with unknown distributions of input features, which makes detection and interpretation difficult.\"},{\"question\":\"What is the XMLAD methodology and how is it structured?\",\"answer\":\"XMLAD is an unsupervised approach that first cleans data by removing outliers, then trains an anomaly detection engine and builds an anomaly classification engine from KPI discretization and predicted labels.\"},{\"question\":\"How does XMLAD produce explainable results?\",\"answer\":\"It uses decision tree classifiers whose decision structures identify the features and thresholds associated with both normal behavior and performance anomalies.\"}]","Explainable Machine Learning for Performance Anomaly Detection and Classification in Mobile Networks | PDF",1785903896,55,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":28},"explainable-machine-learning-for-performance-anomaly-detection-and-classification-in-mobile-networks","",{"@graph":36,"@context":86},[37,54,69],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/explainable-machine-learning-for-performance-anomaly-detection-and-classification-in-mobile-networks/126223/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-23","2026-08-05",true,{"@type":66,"interactionType":67,"userInteractionCount":11},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"Why are anomaly detection and classification challenging in mobile networks?","Question",{"text":76,"@type":77},"They require handling many monitored variables and coping with unknown distributions of input features, which makes detection and interpretation difficult.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"What is the XMLAD methodology and how is it structured?",{"text":81,"@type":77},"XMLAD is an unsupervised approach that first cleans data by removing outliers, then trains an anomaly detection engine and builds an anomaly classification engine from KPI discretization and predicted labels.",{"name":83,"@type":74,"acceptedAnswer":84},"How does XMLAD produce explainable results?",{"text":85,"@type":77},"It uses decision tree classifiers whose decision structures identify the features and thresholds associated with both normal behavior and performance anomalies.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":46,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":46,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]