[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-124060-en":3,"doc-seo-124060-105":30,"detail-sidebar-cat-0-en-105":91},{"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":4,"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},124060,1099514068035,"Ezra","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Machine Learning Approaches in Prediction Problems with Human Activity Patterns - Tesis Doctoral","Human activity analysis using diverse algorithmic approaches, especially Machine Learning methods, has drawn increasing research interest in recent years. Beyond purely human behavior, related domains such as mobility and traffic, consumption forecasting, event attendance, and timetable-driven activities share similar pattern structures and create distinctive predictive challenges. In this thesis, Machine Learning and Deep Learning models are applied to nine regression problems, enhanced by Feature Selection to improve performance and interpretability. Multi-step forecasting and dataset segmentation strategies address longer horizons and transparency needs, with results published in indexed journals.","Programa de Doctorado en Tecnologías de la Información y las  \nComunicaciones  \nMachine Learning Approaches in Prediction Problems with  \nHuman Activity Patterns  \nTesis Doctoral presentada por  \nRICARDO TORRES LÓPEZ  \n2024  \nPrograma de Doctorado en Tecnologías de la Información y las Comunicaciones  \nMachine Learning Approaches in Prediction Problems with  \nHuman Activity Patterns  \nTesis Doctoral presentada por  \nRICARDO TORRES LÓPEZ  \nDirectores:  \nDR. SANCHO SALCEDO SANZ  \nDR. JORGE PÉREZ ARACIL  \nAlcalá de Henares, 2024  \nAbstract  \nThe analysis of human activities using different algorithmic approaches, including ML methods, has attracted the attention of researchers in the last few years. While not exclusively human activities, other activities exhibit patterns like those observed in human behaviour, such as human mobility and traffic patterns, consumption prediction, attendance at events or anyother activity with a timetable.  \nPredictions in these datasets are complex tasks with specific characteristics. For example, the importance of the time tag in which the prediction is performed, the existence of a few exogenous variables to complement the information of the time series, or specific peculiarities and structure for each problem adds extreme difficulty to any prediction task carried out in these datasets. With the complexity of modern problems, it’s essential to have a technique that is adaptable enough to provide precise predictions. Machine Learning approaches have been proved as the only viable solution to achieve this level of flexibility.  \nIn this thesis, different Machine Learning and Deep Learning techniques have been applied to nine regression problems. In addition, to improve the performance of the models and add explainability to the problem, Feature Selection algorithms have been designed and implemented to select the most relevant variables in each problem.  \nMoreover, predicting one time step (one day or one hour ahead, depending on the time granularity) is not always enough for real problems. For instance, in the Emergency Department of Hospitals, where resource allocation is critical, predictions should be developed for longer time horizons. For this reason, we have studied and employed a direct multi-step forecast strategy. This methodology requires the development of single models for each forecasting time step. Thus, a different output for each model is obtained whilst the same input is used.  \nFinally, a data segmentation strategy has been implemented in the last problem addressed in this thesis. It seeks to improve the performance of the predictive models while ensuring transparency in the prediction process. This strategy involves splitting the database into multiple subsets based on various criteria while different models are trained independently on each subset. This approach aims to achieve improved results by training models tailored to the unique characteristics of each subset. Simultaneously, it allows the analysis of the variation of the predictor variable distributions through the subsets and how it affects the model performance. This field of the ML area is known as ensemble regression and has gained popularity in recent years.  \nThe results obtained in this research work have been published in three scientific journals indexed in the Journal Citation Reports (JCR) . A list of the papers related to the research work performed in this PhD thesis can be seen in the Appendix.  \nResumen en Castellano  \nEl análisis de las actividades humanas utilizando diferentes algoritmos, incluidos los métodos de Aprendizaje Automático (Machine Learning), ha atraído la atención de los investigadores en los últimos años. Además existen otras muchas actividades que, sin ser exclusivamente humanas, tienen patrones similares a los observados en el comportamiento humano, como las relacionadas con la movilidad, el tráfico, la predicción de bienes de consumo, asistencia a eventos o cualquierotra actividad lig","cbCaibxiQkNc8FXq","https://ap.wps.com/l/cbCaibxiQkNc8FXq","pdf",23536789,1,200,"English","en",105,"# Abstract\n## Prediction challenges and dataset characteristics\n## Applied ML/DL regression models and feature selection\n## Multi-step forecasting strategy for longer horizons\n## Data segmentation and ensemble regression approach\n## Research outputs and publications","[{\"question\":\"What kinds of prediction problems does the thesis focus on?\",\"answer\":\"The thesis targets regression forecasting problems related to human activity patterns and activity-like behaviors in datasets, such as mobility, traffic, consumption, event attendance, and timetable-based activities.\"},{\"question\":\"How do Feature Selection methods contribute to the models?\",\"answer\":\"Feature Selection algorithms are designed and implemented to choose the most relevant variables for each problem, improving predictive performance and adding explainability.\"},{\"question\":\"Why is a multi-step forecasting strategy used?\",\"answer\":\"Single-step predictions are not sufficient for real scenarios requiring planning over longer horizons, such as resource allocation in hospital emergency departments. A direct multi-step strategy trains models for each future time step using the same input.\"}]","Machine Learning Approaches in Prediction Problems with Human Activity Patterns - Tesis Doctoral | PDF",1785820142,504,{"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":86,"head_meta":88,"extra_data":90,"updated_unix":28},"machine-learning-approaches-in-prediction-problems-with-human-activity-patterns-doctoral-thesis","",{"@graph":36,"@context":85},[37,54,68],{"@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/machine-learning-approaches-in-prediction-problems-with-human-activity-patterns-doctoral-thesis/124060/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04",true,{"@type":65,"interactionType":66,"userInteractionCount":4},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What kinds of prediction problems does the thesis focus on?","Question",{"text":75,"@type":76},"The thesis targets regression forecasting problems related to human activity patterns and activity-like behaviors in datasets, such as mobility, traffic, consumption, event attendance, and timetable-based activities.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How do Feature Selection methods contribute to the models?",{"text":80,"@type":76},"Feature Selection algorithms are designed and implemented to choose the most relevant variables for each problem, improving predictive performance and adding explainability.",{"name":82,"@type":73,"acceptedAnswer":83},"Why is a multi-step forecasting strategy used?",{"text":84,"@type":76},"Single-step predictions are not sufficient for real scenarios requiring planning over longer horizons, such as resource allocation in hospital emergency departments. A direct multi-step strategy trains models for each future time step using the same input.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]