February 1, 2020 Thesis Open Access

digis weldu

Citation Style Language JSON Export

{
  "DOI": "10.20372/nadre:1973", 
  "author": [
    {
      "family": "digis weldu"
    }
  ], 
  "issued": {
    "date-parts": [
      [
        2020, 
        2, 
        1
      ]
    ]
  }, 
  "abstract": "<p>The advent of data-intensive services needs quality Internet services. This in turn, makes Quality<br>\nof Experience (QoE) gain prominent recognition in the telecommunications industry. Ethio telecom<br>\nuses network Quality of Service (QoS) monitoring data obtained from Network Management<br>\nSystems (NMS) tools to comprehend its network performances. However, as QoS measurement<br>\nrefers to network performances, this method does not generally give QoE data as perceived<br>\nby the user. Therefore, QoE estimation models are proposed as solutions in the literature,<br>\nrecently.<br>\nThis study focuses on developing QoE estimation models using QoS features of round-trip time<br>\n(RTT), jitter, loss rate (LR) and throughput, and QoE scores collected using Application for prediCting<br>\nQUality of experience at Interne Access (ACQUA)-based crowdsourcing in Universal<br>\nMobile Telecommunication Systems (UMTS) networks in a real-time basis. Data preparations<br>\ntechniques such as data cleaning and dataset imbalance corrections have been applied to the<br>\ncollected datasets. Machine Learning (ML) algorithms of Articial Neural Network (ANN), KNearest<br>\nNeighbor (KNN) and Random Forest (RF) are selected based on their suitability for multilabel<br>\nproblems. After training these models developed, they are evaluated using commonly used<br>\nperformance metrics such as accuracy, Root Mean Square Error (RMSE) and Receiver Operating<br>\nCharacteristics (ROC).</p>", 
  "title": "Machine Learning Based QoE Estimation Model for Video Streaming over UMTS Network", 
  "type": "thesis", 
  "id": "1973"
}