Sökning: "time-to-event analysis"

Visar resultat 1 - 5 av 16 uppsatser innehållade orden time-to-event analysis.

1. 1. Customer churn prediction in a slow fashion e-commerce context : An analysis of the effect of static data in customer churn prediction

   Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

   Författare :Luca Colasanti; [2023]
   Nyckelord :Survival Analysis; Time To Event prediction; Churn retention; Machine Learning; Deep Learning; Customer Clustering; E-commerce; Analisi di sopravvivenza; Previsione del tempo a evento; Ritenzione dall’abbandono dei clienti; Apprendimento automatico; Apprendimento profondo; Segmentazione della clientela; Commercio elettronico; Överlevnadsanalys; Tid till händelseförutsägelse; Churn Prediction; Maskininlärning; Djuplärning; Kundkluster; E-handel;

   Sammanfattning : Survival analysis is a subfield of statistics where the goal is to analyse and model the data where the outcome is the time until the occurrence of an event of interest. Because of the intrinsic temporal nature of the analysis, the employment of more recently developed sequential models (Recurrent Neural Network (RNN) and Long Short Term Memory (LSTM)) has been paired with the use of dynamic temporal features, in contrast with the past reliance on static ones. LÄS MER

3. 2. Wind Turbine Recovery Forecasting using Survival Analysis

   Master-uppsats, Lunds universitet/Matematisk statistik

   Författare :Anton Palets; [2023]
   Nyckelord :Survival analysis; Recovery Forecast; Wind Turbine; Availability Forecast; AFT model; Aalen s model; Cox regression; Cox Proportional Hazards; Variation Processes; Mathematics and Statistics;

   Sammanfattning : The goal of this thesis is to present a methodology for predicting time until recovery of failed wind turbines. The necessity is motivated by the potential for more accurate wind energy export forecasts. The current approach rests entirely on having an expert examine the turbine and produce a time estimate. LÄS MER

4. 3. Machine learning for risk ranking of component failure : A comparative study of traditional- and survival machine learning approaches applied to historical data

   Uppsats för yrkesexamina på avancerad nivå, Uppsala universitet/Datalogi

   Författare :Fredrik Nilsson; Fanny Fristedt; [2023]
   Nyckelord :Machine learning; Survival analysis; Trains; Freight trains; Damage prediction; Maskininlärning; överlevnadsanalys; tåg; godståg; Skadeprediktion;

   Sammanfattning : This master thesis investigates the use of machine learning for predicting and assessing the risk of railway vehicle component failures. Data used for failure prediction often comes with limitations due to the complex nature of maintenance or sometimes requires investments for the extraction of information. LÄS MER

5. 4. Development of a Machine Learning Survival Analysis Pipeline with Explainable AI for Analyzing the Complexity of ED Crowding : Using Real World Data collected from a Swedish Emergency Department

   Master-uppsats, KTH/Medicinteknik och hälsosystem

   Författare :Tobias Haraldsson; [2023]
   Nyckelord :SHAP; Explainable AI; Survival Analysis; LOS; Machine Learning; ED Crowding; SHAP; Förklarbar AI; Överlevnadsanalys; LOS; Maskininlärning; Överbelastning på Akuten;

   Sammanfattning : One of the biggest challenges in healthcare is Emergency Department (ED)crowding which creates high constraints on the whole healthcare system aswell as the resources within and can be the cause of many adverse events.Is is a well known problem were a lot of research has been done and a lotof solutions has been proposed, yet the problem still stands unsolved. LÄS MER

7. 5. Enhancing failure prediction from timeseries histogram data : through fine-tuned lower-dimensional representations

   Master-uppsats, Högskolan i Halmstad/Akademin för informationsteknologi

   Författare :Vijay Jayaraman; [2023]
   Nyckelord :time-series prediction; time-series histogram analysis; Convolution Neural Network CNN ; Autoencoder; Weibull Time-to-Event WTTE ; Recurrent Neural Network RNN ; Engine turbo charger failure prediciton;

   Sammanfattning : Histogram data are widely used for compressing high-frequency time-series signals due to their ability to capture distributional informa-tion. However, this compression comes at the cost of increased di-mensionality and loss of contextual details from the original features. LÄS MER