The close app is a concept of a mobile application designed to help COVID-19 patients feel less isolated and lonely, therefore, closer to their loved ones with whom they cannot meet. The main principle of the application lies in fulfilling the fundamental needs of the patients and its family identified throughout the research. An isolated patient feels detached from his environment and unfamiliar with the new one, which can be defined as a lack of relatedness. Other essential needs are stimulation - to be distracted and security - to be informed. On the other side, the family members want to feel close with their hospitalized loved ones. The identified needs are relatedness, competence - to know patient’s feelings and state of health and foremost to have a purpose defined as a meaningful connection and knowledge on how to help. The goal of the project is to reduce the feeling of isolation to Covid19 patients by creating an unobtrusive connection with their family and a warm home environment. The project intended to be a rapid answer to the current pandemic and resulted in the isolation of vulnerable patients. The process started from the market research phase, where the main categories of projects, concepts, services, and consumer products were evaluated. The outcome showed that the majority of identified solutions that tackle the problem of closeness are mobile applications. The main principles of interaction derived from the market research are: sharing the same activity, sharing an experience, evoke memories, share memories, develop a relationship. The exploration and ideation had been made based on those outcomes resulted in the creation of three concepts: Evoke memories, Ambient participation (share experience), Active participation (share an activity). Concepts were evaluated throughout the interviews. Additionally, participants were asked about the meaning of closeness. After every iteration, the list of requirements had been updated and led to developing two concepts the mobile application -Digital Closeness and the physical concept - Modular Closeness. The decision needed to be made; two concepts were evaluated using the Harris profile method. Results have shown the possible higher improvement of perceived closeness with faster development and implementation process than the other concept. The Close app focuses on the discreet way of sharing memories which are the source of positive emotions and moments to send short updates to the patient, that both can serve as a distraction for the patient. App provides the patient’s family with the purpose by showing them the importance of creating meaningful content. The patient is also able to send short reactions that will give the family information. Additionally, the Close app offers an ambiance connection that is based on the family location. The familiar picture will be displayed on the patient’s TV during his inactivity, representing the location of the patient’s loved one. Firstly, the concept had been evaluated with the AttrakDiff to assess the hedonic and pragmatic qualities and its attractiveness. Secondly, the evaluation was conducted with hospitalized ex- COVID-19 patients, a family member of the hospitalized patient, and the healthcare professional. The concept had been received significantly positively by all of the stakeholders.

People with diabetes often show symptoms heavily associated with depression, but these symptoms are often caused by the burden of daily diabetes management. The negative feelings caused by this burden are defined as "diabetes burnout''. Some of these negative feelings are caused by social issues. People with diabetes often resort to online sources to find out how to deal with these social issues since health care providers do not focus on this side of diabetes. Furthermore, the social stigma surrounding treatment for mental problems stops people with diabetes from going to a psychologist. In this project, a conversational agent is designed, implemented and evaluated to investigate whether it is capable of reducing social diabetes distress. The agent was designed to give personalized tips based on a social issue the person with diabetes shares. A longitudinal experiment was done over three sessions to evaluate the agent. The results show that the agent is able to reduce the diabetes distress more than a plain textual delivery of tips. The successful application shows the value of conversational agents and provide a basis to deploy such conversational agents in the e-mental health domain. The design we created can be used in future work, where a further personalized approach and a tool measuring the personalization should be investigated in order to better understand why the conversational agent is able to reduce the diabetes distress.

Personality disorders affect 1 in 7 adults reducing their quality of life. Schema-focused therapy (SFT) has become very popular in Psychotherapy in the treatment of personality disorders (PD), unfortunately there is still in increasing societal need for such mental healthcare. Automation in the assessment of SFT allows for Ecological Momentary Assessments (EMA). Resulting in a dynamic assessment of schema-modes and making the treatment more socially available. Automation is realised by Allaart in the form of a conversational agent (CA), but needs a better schema classification algorithm to improve its efficacy. The goal of this study is to evaluate the k-Nearest Neighbour (kNN) algorithm along with Allaart’s dataset. The main question of the study is as follows: How well can a schema be automatically classified from a text using KNN? The method comprises of an experimentalpipeline consisting of 4 stages: Labeling of dataset; pre-processing of the data; schema classification; and evaluation. kNN performed satisfactory in multi-label binary classification with a mean accuracy of 71% and a mean weighted f1-score of 0.62. kNN did not outperform other classification algorithm and performed inadequate in ordinal classification. Results indicate a contrast between majority and minority classes and found a recall of 100% on one of the majority classes. Hence, the data set is concluded to be imbalanced. Due to limitations on the dataset and the CA no reliable conclusion can be made on the performance of kNN in automated schema classification. This study proposed future research to conduct a field experiment such that the CA and its ability to perform EMA is evaluated and reliable data is produced.

In recent years there has been an increase in the number of patients for issues relating to mental illness. To this effect to help with this increase, schema mode assessment through a conversation agent is being used to conduct schema therapy, a form of psychological treatment. To train such an agent, training data labeled by humans is necessary but can be very expensive to conduct. The question being researched is through the use of Active Machine learning is it possible to reduce the amount of required labeled data to do such classification. Three experiments on the use of active learning with currently available classifiers were performed where the active learner attempted to train the classifiers to an accuracy within +/- 3% of the same classifier trained with traditional machine learning on the full data set. The experimental results found that in all cases the use of active learning drastically decreased the number of necessary labeled data the classifier needed to achieve a similar accuracy. Consistently reducing the number by 98% and above answering the initial question. Though possible limitations of the data set and classifiers for such texts may be positively influencing the magnitude of the reduction.

In recent years, there has been an increasing number of patients with mental disorders. A conversational agent is being developed to ensure an easier diagnosis based on the chat between a patient and the agent. The objective of this research is to assess how well Support Vector Machine (SVM) classifies text into its corresponding schema, which are the mental states of the patient. In total, three different classifications have been attempted, Binary, Ordinal, and Per-Questionnaire. The experimental result indicated that SVM is possible to classify 2 out of 7 schema modes, but in general, the performance of SVM was not outperforming with a low f1-score. At the end of the research, SVM was compared to Recurrent Neural Network (RNN) and k-Nearest-Neighbour (kNN) and it turned out that RNN gives the best performance. One of the limitations affecting the result is the quality of the data set. With more correlated labels and a greater size of the data set, improved results can be expected.

Around the world, child helplines through their services provide a safe and confidential space for children to be heard and empowered. The Dutch Kindertelefoon is one of such helplines providing counselling services to children via call and chat all year round. In this thesis, we explore the design of a conversational agent for training counsellors of the Kindertelefoon. More specifically, we explore the design of an agent in a role-play setting where the agent acts as a child help seeker and the user, a counsellor of the helpline. We designed a conversational agent based on the Belief-Desire-Intention (BDI) model of agency that simulates a child victim of school bullying. Through interaction with the agent, a counsellor is able to learn the Five Phase Model, the conversation model that underpins the helpline's counselling methodology to ensure conversations remain child-centered. We tested a prototype based on this design with a group of counsellors at the Kindertelefoon with regards to their counselling self-efficacy and perceived usefulness of the system. Our results show that the conversational agent is able to influence the counselling self-efficacy of users, albeit a decrease in self-efficacy. The opposite would have been preferred for a learning tool to enable counsellors achieve more effective performance over time. However, feedback from participants indicate the potential of this conversational agent as an additional learning opportunity for training counsellors at the helpline.

Schema therapy is a type of psychological treatment for people suffering from personality disorders. A schema is a core psychological state of mind that influences external behaviour through the development of coping styles. Current schema therapy is time inefficient and human-processed. Enabling automatic schema classification helps the overall goal of creating a chatbot that can classify schema modes from conversations. The goal of this research was to optimize a Recurrent Neural Network (RNN) model to classify patient’s schema modes from a dataset containing a recent emotional story from participants and are labeled with SMI questionnaire answers. Three RNN models were created: a binary classification Multilabel RNN, a binary classification Per-Schema RNN and a ordinal classification Per-Schema RNN. The results have shown that the binary classification Multilabel model scores an average F1-score of 0.48. The binary classification Per-Schema model scores an average F1-score of 0.49. While ordinal classification Per-Schema model performs with an average Spearman Coefficient of 0.15.