Sökning: "Maria Movin"

Hittade 2 uppsatser innehållade orden Maria Movin.

  1. 1. Spelling Correction in a Music Entity Search Engine by Learning from Historical Search Queries

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

    Författare :Maria Movin; [2018]
    Nyckelord :Machine Learning; LSTM; Music search; Query Suggestion;

    Sammanfattning : Query spelling correction is an important component of modern search engines that can help users to express their intent, and thus improve search quality. In this study, we investigated with what accuracy a sequence-to-sequence recurrent neural network (RNN) can recognise and correct misspellings in a music search engine, when the model is trained with old search queries. LÄS MER

  3. 2. A Comparison of Resampling Techniques to Handle the Class Imbalance Problem in Machine Learning : Conversion prediction of Spotify Users - A Case Study

    Kandidat-uppsats, KTH/Skolan för datavetenskap och kommunikation (CSC)

    Författare :Michelle Jagelid; Maria Movin; [2017]
    Nyckelord :;

    Sammanfattning : Spotify uses a freemium business model, meaning that it has two main products, one free limited and one premium for paying customers. In this study we investigated machine learning models’ abilities, given user activity data, to predict conversion from free to premium. LÄS MER