Automatic detection of developmental dyslexia from eye movement data
Authors
Date
2019Copyright
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Lukemisen erityisvaikeus eli dysleksia on maailmanlaajuisesti yleisin neurologinen oppimisvaikeus. Se voi hoitamattomana merkittävästi haitata yksilön akateemista menestystä. Erityisvaikeuden tunnistaminen ja hoitaminen aikaisessa vaiheessa voi kuitenkin vähentää huomattavasti häiriön aiheuttamia ongelmia. Tässä tutkimuksessa esitetään menetelmä tunnistaa dysleksia koneoppimisen avulla silmänliikedatasta. Hyödyntämällä suunnittelutieteen periaatteita oli mahdollista saada uutta tietoa käytettyyn aineistoon liittyen sekä luoda koneoppimismalli, joka pystyy luotettavasti tunnistamaan lukemisen erityisvaikeudesta kärsivät henkilöt. Tutkimuksessa käytettiin tukivektorikone- ja satunnaismetsä-menetelmiä ennustavien mallien luomiseksi. Parhaan saadun mallin tunnistamisen yleistarkkuus oli 89,8% ja dyslektikkojen tunnistamisen tarkkuus 75,9%. Dyslexia is the most common neurological learning disability found worldwide. Though it can seriously hinder individuals' academic success, detecting and treating it early on can drastically reduce its negative effect. Detecting dyslexia reliably and with ease is thus of paramount importance. In this thesis, a method using machine learning and eye movement data to predict if the reader has dyslexia is presented. By using the design science approach, it was possible to obtain new information regarding the data used in addition to a model capable of reliably predicting reading disorders. Support Vector Machine and Random Forest were the methods studied and applied to the data. The best model was obtained by the Support Vector Machine classifier using Random Forest to select the most important features: the general accuracy achieved was 89.8% and the accuracy of detecting dyslexics was 75.9%.
Keywords
Metadata
Show full item recordCollections
- Pro gradu -tutkielmat [28158]
Related items
Showing items with similar title or keywords.
-
Detection of developmental dyslexia with machine learning using eye movement data
Raatikainen, Peter; Hautala, Jarkko; Loberg, Otto; Kärkkäinen, Tommi; Leppänen, Paavo; Nieminen, Paavo (Elsevier, 2021)Dyslexia is a common neurocognitive learning disorder that can seriously hinder individuals’ aspirations if not detected and treated early. Instead of costly diagnostic assessment made by experts, in the near future dyslexia ... -
An eye movement study on the mechanisms of reading fluency development
Hautala, Jarkko; Hawelka, Stefan; Ronimus, Miia (Elsevier, 2024)Little is known about how word recognition processes, such as decoding, change when reading fluency improves during the school year. Such knowledge may have practical importance by determining which aspects of reading are ... -
Dual-stage and dual-deficit? Word recognition processes during text reading across the reading fluency continuum
Hautala, Jarkko; Hawelka, Stefan; Aro, Mikko (Springer Science and Business Media LLC, 2022)Central questions in the study of visual word recognition and developmental dyslexia are whether early lexical activation precedes and supports decoding (a dual-stage view) or not (dual-route view), and the locus of deficits ... -
Visual word recognition in fluent and dysfluent readers in the transparent Finnish orthography
Hautala, Jarkko (University of Jyväskylä, 2012) -
What information should I look for again? : Attentional difficulties distracts reading of task assignments
Hautala, Jarkko; Loberg, Otto; Azaiez, Najla; Taskinen, Sara; Tiffin-Richards, Simon P.; Leppänen, Paavo H.T. (Elsevier, 2019)This large-scale eye-movement study (N = 164) investigated how students read short task assignments to complete information search problems and how their cognitive resources are associated with this reading behavior. These ...