In this thesis we examine the question whether it is possible to model approach and avoidance behaviour with probabilistic machine learning. The results from this project will primarily aid in our collective understanding of human existence. Secondly, it will extend the knowledge with regards to probabilistic machine learning in the Neuroscience domain. We aid this through building a Variational Recurrent Neural Network (VRNN) that is trained on Electroencephalography (EEG)-data from participants that is subjected to odours with varying pleasantness. The pleasantness of the odours is used to divide the participants into two classes based on their self reported experience. This data is used to train the VRNN. The performance of the VRNN is evaluated by how well we are able to reconstruct the original data from a low dimensional latent representation. In this task the model performs on a similar level as related works. We further investigate how changes in the latent space effects reconstructed data. Despite being disentangled, the latent variables are hard to interpret. Furthermore we try to classify and cluster the latent space as either approach or avoidance behaviour with a Support Vector Machine and Uniform Manifold Approximation. The classification results are only slightly better than random, indicating that the learned latent space is not suitable for the task This is most likely due to the patterns that make up approach and avoidance behaviour is seen as noise by the VRNN. This leads to the patterns not being accurately modelled. This is shown by the evidence that frontal α -asymmetry that exists in the data is not reconstructed by the model. The conclusion is therefore that a VRNN is less suitable for modelling underlying behaviour from raw EEG data due to the low signal to noise ratio. We instead suggests to focus on specific frequency ranges in specific regions when applying machine learning in this domain. / Den här uppsatsen behandlar frågan huruvida det är möjligt att modellera närmande och frånstötande beteendemönster med hjälp av maskininlärning. Resultaten från detta projekt ämnar huvudsakligen att främja vidare förståelse av den mänskliga existensen. Vidare ämnar den även att utvidga förståelsen av hur probabilistisk maskininlärning kan användas för att utforska dylika hänseenden. Vi genomför detta genom att bygga en Variational Recurrent Neural Network-modell (VRNN) som tränas på data från experiment där personer utsätts för olika lukter samtidigt som deras Elektroencefalografi (EEG) spelas in. Deltagarna delas in i två klasser beroende på deras självrapporterade upplevelse av luktens njutbarhet. Maskininlärningsmodellen utvärderas genom att vi analyserar hur väl den lyckas rekonstruera datan. Detta lyckas den väl med. Vidare så undersöker vi hur förändringar i modellens latenta rum påverkar rekonstrueringen av datan. Resultaten från det experimentet är ej tydliga. Vidare så försöker vi klassificera och klustra det latenta rummet med avseende på närmande och frånstötande beteende med hjälp av en Support Vector Machine och Uniform Manifold Approximation. Resultaten från dessa experiment är att vi inte lyckas klassificera eller klustra det latenta rummet med avseende på närmande och frånstötande beteende bättre än slumpen. Vi argumenterar för att detta beror på att de underliggande mönster som skapar dessa beteenden ses som brus av VRNN-modellen och därmed inte modelleras. Detta visas genom att frontal α-asymmetri som existerar i datan ej rekonstrueras av modellen. Slutsaten blir därmed att en VRNN är mindre passande att använda vid modellering av underliggande beteenden av obehandlad EEG data. Detta på grund av det låga signal till brus-förhållandet i EEG-datan. Vi föreslår att istället fokusera på specifika frekvensområden i specifika hjärnregioner när maskininlärning appliceras på EEG.

This thesis focuses on statistical and machine learning methods designed for sequential and repeated measurement data. We start off by considering the classic general linear model (MANOVA) followed by its generalization, the growth curve model (GMANOVA), designed for analysis of repeated measurement data. By considering a binary classification problem of normal data together with the corresponding maximum likelihood estimators for the growth curve model, we demonstrate how a classification rule based on linear discriminant analysis can be derived which can be used for repeated measurement data in a meaningful way. We proceed to the topics of neural networks which serve as our second method of classification. The reader is introduced to classic neural networks and relevant subtopics are discussed. We present a generalization of the classic neural network model to the recurrent neural network model and the LSTM model which are designed for sequential data. Lastly, we present three types of data sets with an total of eight cases where the discussed classification methods are tested. / Den här uppsatsen introducerar klassificeringsmetoder skapade för data av typen upprepade mätningar och sekventiell data. Den klassiska MANOVA modellen introduceras först som en grund för den mer allmäna tillväxtkurvemodellen(GMANOVA), som i sin tur används för att modellera upprepade mätningar på ett meningsfullt sätt. Under antagandet av normalfördelad data så härleds en binär klassificeringsmetod baserad på linjär diskriminantanalys, som tillsammans med maximum likelihood-skattningar från tillväxtkurvemodellen ger en binär klassificeringsregel för data av typen upprepade mätningarn. Vi fortsätter med att introducera läsaren för klassiska neurala nätverk och relevanta ämnen diskuteras. Vi generaliserar teorin kring neurala nätverk till typen "recurrent" neurala nätverk och LSTM som är designade för sekventiell data. Avslutningsvis så testas klassificeringsmetoderna på tre typer av data i totalt åtta olika fall.

No description available.

This study aims to investigate the implementation of an AI model that predicts customer purchases, in the telecom industry. The thesis also outlines how such an AI model can assist decision-making in marketing strategies. It is concluded that designing the AI model by following a Recurrent Neural Network (RNN) architecture with a Long Short-Term Memory (LSTM) layer, allow for a successful implementation with satisfactory model performances. Stepwise instructions to construct such model is presented in the methodology section of the study. The RNN-LSTM model further serves as an assisting tool for marketers to assess how a consumer’s website behavior affect their purchase behavior over time, in a quantitative way - by observing what the authors refer to as the Customer Purchase Propensity Journey (CPPJ). The firm empirical basis of CPPJ, can help organizations improve their allocation of marketing resources, as well as benefit the organization’s online presence by allowing for personalization of the customer experience. / Denna studie undersöker implementeringen av en AI-modell som förutspår kunders köp, inom telekombranschen. Studien syftar även till att påvisa hur en sådan AI-modell kan understödja beslutsfattande i marknadsföringsstrategier. Genom att designa AI-modellen med en Recurrent Neural Network (RNN) arkitektur med ett Long Short-Term Memory (LSTM) lager, drar studien slutsatsen att en sådan design möjliggör en framgångsrik implementering med tillfredsställande modellprestation. Instruktioner erhålls stegvis för att konstruera modellen i studiens metodikavsnitt. RNN-LSTM-modellen kan med fördel användas som ett hjälpande verktyg till marknadsförare för att bedöma hur en kunds beteendemönster på en hemsida påverkar deras köpbeteende över tiden, på ett kvantitativt sätt - genom att observera det ramverk som författarna kallar för Kundköpbenägenhetsresan, på engelska Customer Purchase Propensity Journey (CPPJ). Den empiriska grunden av CPPJ kan hjälpa organisationer att förbättra allokeringen av marknadsföringsresurser, samt gynna deras digitala närvaro genom att möjliggöra mer relevant personalisering i kundupplevelsen.

在公司財務的領域中，探討公司資本結構決策主要有三個主流理論：靜態抵換理論、融資順位理論以及折時理論。本篇文章採用重複事件分析法，首先沿用Baker and Wurgler (2002)中提及之五個因素做為自變數，研究影響公司辦理現金增資危險函數之因子研究，研究結果顯示，公司現金增資之危險函數與財務槓桿成正向關係，此項證據傾向支持融資順位理論，然而本篇論文研究結果，並無顯著證據支持折時理論。本篇論文接著建立另一組變素設定，將價格趨勢納入模型中，取代原來在Baker and Wurgler(2002)中觀察折時現象之因子，結果顯示折時現象是顯著的。因此，本篇論文研究結果並未對是否支持折時理論下定論，值得思考的是，欲觀察公司是否存在折時現象，除了Baker and Wurgler(2002)中提及之變數之外，直接將價格趨勢納入模型或許是另一個可行之道。 / In the field of traditional corporate financing theories, there are three mainstream theories leading the way while talking about the firms’ financing decisions: static trade-off theory, pecking order theory, and market timing theory. In this paper, we apply the recurrent event analysis and follow the independent variables appearing in the Baker and Wurgler (2002) first to examine the factors that affect firms’ hazard rate to offer seasoned equity. The results indicate that higher leverage is in positive relation with the hazard rate of firms’ seasoned equity offering, meaning that firms’ financing decisions follow the pecking order theory to some degree. However, while the recurrent event analysis is adopted, the market timing effect becomes insignificant when considering the independent variables appearing in the Baker and Wurgler(2002). As a result, we proceed to establish another set of covariates in which the price trend factor is involved to examine the market timing effect. While the price trend factor is substituted for the market-to-book ratio to represent the market timing effect, the market timing effect turns out to be significant. Thus, we consider that using the price trend of the market directly may be a suitable way to examine the market timing effect.

This thesis focuses on the effects of the human papillomavirus (HPV) in tonsillar cancer, mobile tongue cancer, and recurrent respiratory papillomatosis (RRP). The purpose was to characterize patients with RRP in northern Sweden in order to identify more care-intensive RRP patients and to describe the voice and quality of life aspects that follow RRP. Further aims were to confirm the expected increase of HPV-positive tonsillar cancer cases in northern Sweden, and to study the correlation between HPV, its surrogate marker p16 and HPV receptor syndecan-1 in both tonsillar cancer and mobile tongue cancer. A total of 27 consecutive patients with RRP were evaluated at 3 months postoperatively using the voice handicap index (VHI) and SF-36 questionnaires to assess the impact on life and voice in a RRP population. The values were compared to normative data. This report was further extended by examining consecutive data from 21 new patients in order to characterize RRP patients in northern Sweden. In order to study HPV DNA in tonsillar (n= 65) and mobile tongue cancer (n=109), HPV DNA was extracted from paraffin-embedded biopsies and detected by polymerase chain reaction using general primers Gp5+/6+ and CpI/IIG. Expression of HPV surrogate marker p16 and the HPV receptor syndecan-1 was analysed by immunohistochemistry. Patients that underwent more than one RRP surgery per year were younger than those treated less frequently and they had significantly impaired voice quality as compared to normal subjects. Females, patients with frequent surgical treatment sessions, and patients with the high-risk HPV subtypes scored significantly lower in several domains of the quality of life assessment as compared with normal subjects. Forty-eight RRP patients had a median age of 44.5 years; 71% were men and 29% females, preferentially infected with HPV6. Patients with high surgical treatment frequency/year showed more widespread RRP in the larynx compared to the patients treated less frequently. A total of 214 tonsillar cancer cases were identified. The vast majority were men. They had a median age of 58 years at diagnosis and expressed HPV as well as p16. The incidence of tonsillar cancer revealed a 2,7-fold increase in men between the years 1990 and 2013. The study demonstrates a strong association between p16 and HPV infection in tonsillar malignancies. These findings are in contrast to the mobile tongue cancer cases, where no evidence of HPV DNA could be detected although one-third showed p16 staining. This demonstrated a poor correlation between HPV and p16 in mobile tongue cancer. There was no difference in the expression of the primary HPV receptor, syndecan-1, between tonsillar and mobile tongue cancer. In conclusion, the frequency of RRP operations, age at onset, gender and subtype of the HPV may be used as factors to predict voice disability. RRP patients with high surgical treatment frequency were significantly younger and had a more widespread laryngeal disease compared to the low-frequency treated group. This study confirms the existence of a clinical RRP group, not primarily related to HPV subtype, but to a more care-intensive RRP population. Our findings identify a 2,7-fold increase in the incidence of tonsillar cancer, HPV and p16 in men between 1990-2013. We can use p16 to detect HPV in tonsillar cancer but not in tongue cancer. The introduction of vaccination against HPV may have a role in the prevention of specific HPV-subtype positive head and neck malignancies and recurrent respiratory papillomatosis since the current vaccine protects against HPV6, 11, 16, 18, 31, 33, 45, 52 and 58. Males will definitely benefit indirectly from vaccination of females, though males will still remain at risk of cancers associated with HPV. This highlights the need for sex-neutral vaccination strategy. Our intention is that this thesis will provide scientific data to support a gender-neutral vaccination and to develop simple tools to detect HPV in tonsillar cancer. / Syftet med avhandlingen är att beskriva effekterna av humant papillomvirus (HPV) vid cancer i halsmandlarna, cancer i tungan och vid luftvägspapillom. Totalt 27 patienter med luftvägspapillom (RRP) under åren 2004-2012 utvärderades 3 månader efter operationen med röst handikapp index (VHI) och livskvalitetformuläret SF-36. Resultaten jämfördes med normal data. Studiematerialet utökades med 21 patienter till totalt 48 RRP patienter i syfte att karakterisera patientgruppen i norra Sverige. För att studera HPV-DNA i tonsillcancer (n = 65) och i cancer i mobil del av tungan (n = 109) extraherades HPV-DNA från paraffininbäddade provbitar som sedan analyserades med PCR teknik och GP5 + / 6 + och CPI/IIG primer. Uttryck av surrogatmarkör p16 och HPV-receptorn syndekan -1 analyserades med immunhistokemi. RRP patienter hade en medianålder på 44,5 år; 71% var män och 29% kvinnor, företrädesvis infekterade med HPV6. Patienter som opererades mer än en gång per år var yngre än de som behandlats mindre ofta och hade en statistiskt sämre röstkvalitet än friska kontroller. Kvinnor, patienter med täta kirurgiska behandlingsintervall och högrisk-HPV hade signifikant sämre livskvalitet jämfört med friska kontroller. Patienter med hög kirurgisk behandlingsfrekvens per år var signifikant yngre och hade mer utbredd RRP sjukdom i luftstrupen, jämfört med gruppen med låg behandlingsfrekvens. Sammanlagt, 214 fall av halsmandelscancer identifierades i norra Sverige under åren 1990-2013; majoriteten var män, med en medianålder på 58 år och positiva för både HPV och p16. Andelen halsmandelscancer fall ökade med 2,7 gånger bland männen på 23 år. Vi fann ett starkt samband mellan uttryck av p16 och HPV infektion i halsmandelscancer men inte i HPV-negativ, delvis p16-positiv (33%) mobil tungcancer. Det fanns ingen skillnad i uttrycket av den primära HPV-receptorn, syndekan -1, jämförande tung-, och halsmandelscancer. Antalet RRP operationer, ålder vid insjuknandet, kön och genetisk variant av HPV kan användas som indikatorer för att förutsäga grad av röststörning. RRP patienter med hög kirurgisk behandlingsfrekvens var signifikant yngre och hade en mer utbredd luftvägssjukdom jämfört med RRP patienter som behandlas mindre ofta. Vi har identifierat en undergrupp av RRP patienter som inte primärt karakteriseras efter HPV virusets genetik utan av ett mer vårdintensivt förlopp. Den aktuella avhandlingen har identifierat en 2,7-faldig ökning av antalet halsmandelscancer hos män och ett starkt samband mellan p16 och HPV infektion i halsmandlar men inte i HPV-negativ tungcancer som inte korrelerar till p16 uttryck. Vi kan använda p16 för att påvisa HPV i tonsillcancer men inte i cancer i mobil tunga. Idag ingår HPV vaccination i det allmänna vaccinationsprogrammet för flickor. Vi förväntar oss en tydlig profylaktisk effekt avseende insjuknande i HPV-relaterad huvud- och hals cancer samt luftvägspapillom eftersom vaccinet skyddar mot HPV bl.a. 6, 11, 16 och 18. Män kommer definitivt att gynnas indirekt genom vaccination av kvinnor men kommer att ha fortsatt högre risk än kvinnor att insjukna i HPV relaterad cancer vilket understryker behovet av könsneutral vaccination. Vår avsikt med avhandlingen är att ge vetenskapligt stöd för könsneutralt vaccination och enkla metoder att påvisa halsmandelscancer.

Among adolescents, the day-to-day functioning of the hypothalamo-pituitary-adrenal-axis (HPA-axis) and of the autonomic nervous system (ANS) and their relationships with stress, subjective health complaints and psychological factors such as self-esteem, studied in naturalistic settings, have been largely unexplored. This thesis aimed to investigate the diurnal activity of the HPA-axis (Studies I &amp; II) in terms of salivary cortisol and the ANS/SNS system (Study III) in terms of salivary alpha-amylase (sAA) in mid-adolescent girls and boys. Additionally, linkages between self-reported stress, self-esteem, recurrent pain and biomarkers were investigated. A further aim was to describe potential differences between girls and boys respectively. Study I showed that both girls and boys exhibited the typical diurnal cortisol profile with high levels in the morning that decreased throughout the day. Girls had higher total cortisol levels, while no differences emerged for measures of the cortisol increase. Study II showed no significant linkages between self-ratings of stress and cortisol. However, stress was associated with recurrent pain in girls. Study III showed that, for girls, both self-esteem and self-reported stress were related to morning levels of both cortisol and sAA, to the diurnal sAA output and to a conjoint measure of amylase over cortisol, AOC. To conclude, the findings suggest that both stress and self-esteem may be linked to different measures of ANS and HPA-axis activity, but also to measures of ANS and HPA-axis dysregulation, particularly among mid-adolescent girls. / <p>At the time of the doctoral defense, the following papers were unpublished and had a status as follows: Paper 2: Manuscript. Paper 3: Manuscript.</p>

Cette thèse porte sur les modèles génératifs en apprentissage automatique. Deux nouveaux modèles basés sur les réseaux de neurones y sont proposés. Le premier modèle possède une représentation interne où une certaine structure a été imposée afin d’ordonner les caractéristiques apprises. Le deuxième modèle parvient à exploiter la structure topologique des données observées, et d’en tenir compte lors de la phase générative. Cette thèse présente également une des premières applications de l’apprentissage automatique au problème de la tractographie du cerveau. Pour ce faire, un réseau de neurones récurrent est appliqué à des données de diffusion afin d’obtenir une représentation des fibres de la matière blanche sous forme de séquences de points en trois dimensions.

Implantation requires complex molecular and cellular events involving coagulation, angiogenesis and immunological processes that need to be well regulated for a pregnancy to establish and progress normally.  The overall aim of this thesis was to study different models associated with atypical angiogenesis, impaired implantation and/or placentation, such as recurrent miscarriage (RM), oocyte donation (OD) and preeclampsia. Histidine-rich glycoprotein (HRG), a serum protein with angiogenic potential has been previously shown to have an impact on implantation and fertility.  In two retrospective case-control studies, women suffering from RM (Study I) and gestational hypertensive disorders (GHD) (Study IV) have been compared to healthy control women, regarding carriership of HRG genotypes (HRG A1042G and C633T SNP, respectively).  According to the findings of this thesis, heterozygous carriers of the HRG A1042G SNP suffer from RM more seldom than homozygous carriers (Study I).  Additionally, the presence of the HRG 633T allele was associated with increased odds of GHD (GHD IV).  Studies II and III comprised a national cohort of relatively young women with optimal health status conceiving singletons with donated oocytes versus autologous oocytes (spontaneously or via IVF).  We explored differences in various obstetric (Study II) and neonatal (Study III) outcomes from the Swedish Medical Birth Register.  Women conceiving with donated oocytes had a higher risk of GHD, induction of labor and cesarean section, as well as postpartum hemorrhage and retained placenta, when compared to autologously conceiving women.  OD infants had higher odds of prematurity and lower birthweight and length when born preterm, compared to neonates from autologous oocytes.  With regard to the indication of OD treatment, higher intervention but neverthelss favourable neonatal outcomes were observed in women with diminished ovarian reserve; the risk of GHD did not differ among OD recipients after adjustment. In conclusion, HRG genetic variation appears to contribute to placental dysfunction disorders.  HRG is potential biomarker that may contribute in the prediction of the individual susceptibility for RM and GHD.  Regarding OD in Sweden, the recipients-despite being of optimal age and health status- need careful preconceptional counselling and closer prenatal monitoring, mainly due to increased prevalence of hypertensive disorders and prematurity.

The purpose of the present study was to examine the role that specific factors play in the quality of life (QL) for women with recurrent breast cancer.