Browsing by Subject "syväoppiminen"
Now showing items 1-4 of 4
-
(2022)This master's thesis seeks to conceptually replicate psychologist Michael Kosinski's study, published in 2021 in Nature Scientific Reports, in which he trained a cross-validated logistic regression model to predict political orientations from facial images. Kosinski reported that his model achieved an accuracy of 72\%, which is significantly higher than the 55\% accuracy measured in humans for the same task. Kosinski's research attracted a huge amount of attention and also accusations of pseudoscience. Where Kosinski trained his model with facial features containing information for example about head position and emotions, in this thesis I use a deep learning convolutional neural network for the same task. Also, I train my model with Finnish data, consisting of photos of the faces of Finnish left- and right-wing candidates gathered from the 2021 municipal elections. I research whether a convolutional neural network can learn to predict from candidates' faces whether a member of a Finnish party belongs to either the right-wing Coalition Party (Coalition) or the left-wing Left Alliance (Left Alliance) with better than 55\% accuracy, and what is the possible role of color information on the classification accuracy of the model. On this basis, I also consider the wider ethical issues surrounding these types of models and the technological advances they bring. There has been a recent ethical debate on the widespread use of facial recognition technology in relation to issues such as human autonomy, privacy, and civil liberties. In the context of previous scientific findings, there has also been debate about the potential ability of facial recognition technologies to reveal information about our most personal traits, such as sexual orientation, personality, and emotional states. Thus, facial recognition technologies are also closely related to privacy issues. In his original article, Michael Kosinski did not underestimate the many problematic ethical issues that the use of facial recognition technology can raise. He did, however, underline the role of science in trying to determine the function, capability, and accuracy of these technologies. Only through research can we gain insights into these technologies, which can then potentially be used to inform societal decision-making. This research approach is also the aim of this Master's thesis.
-
Koirien ja kissojen suolistoloisten tunnistaminen ulostenäytteistä Vetscan Imagyst-analysaattorilla (2024)Koirilla ja kissoilla esiintyy useita suolistoloisia, jotka voivat aiheuttaa isännälleen terveyshaittoja tai muodostaa ihmisille zoonoosiriskin. Suomessa rutiiniloishäätöjä parempi lähestymistapa monessa tapauksessa on tutkia ulostenäyte ja lääkitä eläin vain tarvittaessa. Siksi tarve kustannustehokkaalle, luotettavalle, herkälle ja helppokäyttöiselle menetelmälle loisten osoittamiseksi ulosteista on ilmeinen. Pitkään käytössä olleiden (useimmiten flotaatioon perustuvien) menetelmien rinnalle on viime aikoina kehitetty tekoälyä hyödyntäviä laitteistoja, jotka voivat vastata osaltaan tähän tarpeeseen. Niiden etuna perinteisiin manuaalisiin menetelmiin verrattuna on näytteen analysoijasta johtuvan vaihtelun poistaminen sekä nopea ja standardoitu näytteenkäsittely. Tutkimuksessa selvitettiin miten hyvin tekoälyä ja syväoppimista hyödyntävä Vetscan Imagyst -analysaattori löytää ja tunnistaa koirien ja kissojen ulostenäytteistä sisäloisten munia ja kystamuotoja ja verrattiin laitteen antamia tuloksia yleisesti käytössä olevien manuaalisten flotaatiomenetelmien antamiin tuloksiin. Oletuksena oli, että Vetscan Imagyst -menetelmä pystyisi tunnistamaan näytteistä suolistoloisten munat/kystamuodot yhtä hyvin kuin manuaaliset menetelmät. Näytteenä käytettiin Yliopistollisen eläinsairaalan keskuslaboratorioon (YESLAB) lähetettyjä kissojen ja koirien ulostenäytteitä, joista oli tilattu ulosteen sisäloistutkimus flotaatiomenetelmällä (n=42). Tutkimukset tehtiin loka-joulukuussa 2021. Näytteet tutkittiin YESLAB:ssa käytössä olevalla muunnoksella passiivisesta flotaatiomenetelmästä (modifioitu passiivinen flotaatiomenetelmä), sentrifugaatioflotaatiolla parasitologian laboratoriossa sekä Vetscan Imagyst -analysaattorilla. Sekä modifioidussa passiivisessa flotaatiomenetelmässä että sentrifugaatioflotaatiomenetelmässä flotaatioliuoksena käytettiin magnesiumsulfaattia. Menetelmien herkkyyttä tutkittiin etukäteen Toxocara-munilla ympätyillä näytteillä. Tutkimuksessa löytyi kaksi Toxocara-, yksi Trichuris-, yksi Capillaria- ja 13 kokkidialkueläinpositiivista näytettä. Vetscan Imagyst tunnisti kaikki positiiviset näytteet, joskin se nimesi Capillarian väärin Trichurikseksi ja kaikki kokkidit Cystoisosporiksi vaikka yhtä näytettä lukuun ottamatta kaikki havaitut kokkidit olivat Eimerioita. Lisäksi Vetscan Imagyst -menetelmällä tuli kolme virhepositiivista löydöstä. Sentrifugaatioflotaatiomenetelmä ei löytänyt Trichuris-positiivista näytettä eikä yhtä kokkidipositiivista näytettä. Modifioitu passiivinen flotaatiomenetelmä toimi heikoiten ja jätti löytämättä kahdeksan loispositiivista näytettä. Modifioidulla passiivisella flotaatiomenetelmällä tai sentrifugaatioflotaatiomenetelmällä ei tullut virhepositiivisia tuloksia. Kun menetelmien herkkyyttä arvioitiin etukäteen ympätyillä näytteillä, sentrifugaatioflotaatiomenetelmä löysi määrällisesti enemmän Toxocara canis munia kuin Vetscan Imagyst. Vetscan Imagyst on helppo- ja nopeakäyttöinen ja se erottaa hyvin näytteissä olevat loisten munat ja kystamuodot. Syväoppimista hyödyntävänä menetelmänä Vetscan Imagyst tulee todennäköisesti tunnistamaan havaitut löydökset jatkossa entistä paremmin algoritmien tarkennuttua.
-
(2022)Automatic question answering and question generation are two closely related natural language processing tasks. They both have been studied for decades, and both have a wide range of uses. While systems that can answer questions formed in natural language can help with all kinds of information needs, automatic question generation can be used, for example, to automatically create reading comprehension tasks and improve the interactivity of virtual assistants. These days, the best results in both question answering and question generation are obtained by utilizing pre-trained neural language models based on the transformer architecture. Such models are typically first pre-trained with raw language data and then fine-tuned for various tasks using task-specific annotated datasets. So far, no models that can answer or generate questions purely in Finnish have been reported. In order to create them using modern transformer-based methods, both a pre-trained language model and a sufficiently big dataset suitable for question answering or question generation fine-tuning are required. Although some suitable models that have been pre-trained with Finnish or multilingual data are already available, a big bottleneck is the lack of annotated data needed for fine-tuning the models. In this thesis, I create the first transformer-based neural network models for Finnish question answering and question generation. I present a method for creating a dataset for fine-tuning pre-trained models for the two tasks. The dataset creation is based on automatic translation of an existing dataset (SQuAD) and automatic normalization of the translated data. Using the created dataset, I fine-tune several pre-trained models to answer and generate questions in Finnish and evaluate their performance. I use monolingual BERT and GPT-2 models as well as a multilingual BERT model. The results show that the transformer architecture is well suited also for Finnish question answering and question generation. They also indicate that the synthetically generated dataset can be a useful fine-tuning resource for these tasks. The best results in both tasks are obtained by fine-tuned BERT models which have been pre-trained with only Finnish data. The fine-tuned multilingual BERT models come in close, whereas fine-tuned GPT-2 models are generally found to underperform. The data developed for this thesis will be released to the research community to support future research on question answering and generation, and the models will be released as benchmarks.
-
(2023)It has been observed that children’s interest towards natural sciences decreases as they grow up and start middle school. The decrease of interest towards natural sciences and studying them has led to a situation, where science and technology students’ relative share of higher education students has been falling in some of the OECD countries. Formal education needs support to carry out fascinating science education for children and to maintain their interest towards natural sciences. One of the opportunities to fascinate children towards sciences are science centers and their activities either combined with formal education or organized in an informal form like science camps. The aim of this research is to survey the biological knowledge, the interest towards biology and the ability to apply the biological knowledge of science campers (mainly 5th and 6th graders) and to study the connections between them. At the same time, the effect of age and gender of the science campers is studied. To survey the biological knowledge of the campers, multiple-choice test was used and the fascination with biology was investigated with Likert scale variables. The ability to apply the biological knowledge was investigated with drawings that were produced during a biology-themed science camp program. The data was analyzed using quantitative methods. The results indicated that the science campers were relatively fascinated with biology and that they master biological knowledge quite well but the ability to apply the knowledge was varying. There were no differences between different ages nor genders but there are many factors such as the economic status of campers’ families that might have affected on the results. There was found a statistically significant connection between biological knowledge and fascination with biology, which is not surprising, but it speaks in favor of the importance of supporting fascination in terms of learning.
Now showing items 1-4 of 4