Anas Tukur Balarabe, Ivan Jordanov

While fine-tuning a transfer learning model alleviates the need for a vast amount of training data, it still comes with a few challenges. One of them is the range of image dimensions that the input layer of a model accepts. This issue is of interest, especially in tasks that require the use of a transfer learning model. In scene classification, for instance, images could come in varying sizes that could be too large/small to be fed into the first layer of the architecture. While resizing could be used to trim images to a required shape, that is usually not possible for images with tiny dimension s, for example, in the case of the EuroSAT dataset. This paper proposes an Xception model-based framework that accepts images of arbitrary size and then resizes or interpolates them before extracting and enhancing the discriminative features using an adaptive dilation module. After applying the approach for scene classification problems and carrying out a number of experiments and simulations, we achieved 98.55% accuracy on the EuroSAT dataset, 99.22% on UCM, 96.15% on AID and 96.04% on the SIRI-WHU dataset, respectively. We also monitored the micro-average and macro-average ROC curve scores for all the datasets to further evaluate the proposed model’s effectiveness.

Elizabeth A. Matthews, Irina Koleva, Sujana Basnet

Video games are a popular topic in academia, encompassing many research subtopics. Often researchers investigating video game subjects partition the participants by the participant’s familiarity with video games. Unlike other demographic categories on which research subjects are divided, such as age range or income, sorting participants into gamer skill categories, such as “Expert Gamer”, has no objective consensus on what the labels should be. The inconsistency in terminology makes comparison between research works impossible. Hours per week is often a standard but does not provide statistical significance with many factors and is not as objective as it may seem. This paper highlights this problem, collects demographic questions in existing research, and showcases the data collected from a user study with these questions. The results show that self-chosen categories are a statistically significant metric and is recommended as an easy-to-obtain value.

J. David Patón-Romero, Ricardo Vinuesa, Letizia Jaccheri, Maria Teresa Baldassarre

When talking about sustainability, we usually think that it is only about safeguarding the environment; nothing is further from reality. Of course, the environment is a crucial component of sustainability and our survival, but it is important to recall that the society and the economy play important roles in this regard, and without the interconnection and development of t hese three perspectives it will not be possible to achieve sustainable progress. The Sustainable Development Goals (SDGs) established by the United Nations (UN) defend this idea and address the main challenges that humanity faces. One of these challenges is gender equality, which is identified in the perspective of social sustainability through SDG 5. Gender equality is a very complex and difficult challenge to address due to the great cultural diversity of our society. Thus, achieving this goal will requir e laying a solid foundation and working together by combining very different fields of knowledge. In this sense, Artificial Intelligence (AI) is one of the fields that is currently having the greatest impact and relevance for the development of new technologies and for the advancement of numerous areas. This growing evolution of AI demonstrates that its repercussions at the social level must be analyzed and addressed in such a way that AI becomes a positive asset for sustainability and, in this particular case, for gender equality. For all these reasons, this study aims to analyze the current state of the art and collect the existing knowledge in the fields of AI and gender equality, by conducting a Systematic Mapping Study (SMS). The obtained results and findings have allowed us to identify the most relevant advances in this regard, as well as the gaps and drawbacks that currently exist and on which we must urgently focus to address gender equality both in and by AI. In the same way, these find ings demonstrate the limited joint development of both fields, but also indicate an increase in the relevance and the number of proposals that these fields are receiving in recent years.

Daisuke Fukui, Takushige Katsura, Masashi Egi, Norihisa Komoda, Takenao Ohkawa

We previously proposed a method for estimating pleasant and unpleasant emotions with high accuracy using only total hemoglobin data measured with a lightweight functional near infrared spectroscopy device. In this study , we u sed the method to evaluate the accuracy of estimating 20 types of emotions selected as uniformly distributed emotions in Russell’s circumplex model. We first divided the 20 types of emotions into four groups, corresponding to the four quadrants of Russell’s circumplex model and evaluated the estimation accuracy of each quadrant. The results indicate that the activation quadrant was estimated with high accuracy when the emotion was strongly aroused, with 76.7% recall for the pleasant–activation quadrant and 72.2% recall for the unpleasant–activation quadrant. We then evaluated the estimation accuracy of the 20 emotions individually. The results indicate that “excited” and “lethargic” were estimated with high accuracy, with 73.3% recall for “excited” and 61.5% recall for “lethargic,” and recall of “excited” improved to 80% when the emotion was strongly aroused. The results of this study indicate that the more strongly emotions included in activation quadrant in Russell’s circumplex model are aroused, the more accurately they can be classified. “Excited” and “lethargic” could be estimated with high accuracy regardless of the degree of emotional arousal.

Päivi Ovaskainen, Reima Suomi, Pirkko Nykänen

Rationale, aims and objectives: Resources to deliver healthcare are getting scarce all around the world. Clinical pathways are key tools to make health delivery in clinical settings efficient. C linical pathway design, operation and follow -up all suffer from bad information systems integration. This might badly harm the adherence to clinical pathways. W e have very little research on how clinical pathway s are followed. This stu dy assesses the functioning of a clinical pathway for elderly people in a Finnish case. At the same time it documents the hardships of getting information on the patient flows in a clinical pathway. Method: To support our theoretical discussion with a real case, a case study was performed on the follow- up of clinical pathway for elderly patients in emergency care. The data is on all patients aged 75 years or more who had visited the emergency unit of the city hospital in 2006–2008 was collected. The study sample comprised 24,195 admissions. The flow of patients after the first care at the emergency unit was analyzed, collecting data from various scattered sources. Results: The results show that the planned patient pathways seem not to be followed always. Yearly about 17% of the patients were referred to the university hospital, the more expensive care, but about 74% of these referrals did not follow the agreed procedure. The excess costs of referring patients to a non-standard, more expensive clinical pathway were 2,57 Million Euros over a period of three years for the studied population of 24,195 admissions, also averaging 106 Euros per admission. The case study clearly documents that adherence to the clinical pathway in the case is not as planned, and that system integration difficulties severely harm endeavors to analyze the functioning of and adherence to the pathway. Conclusion: Clinical pathways were not followed in the case as planned and expected. Bad data because of missing system integration made the follow -up of the clinical pathway adherence in our very challenging. Follow-up information of clinical pathway flows is often hard to collect because of fragmented information systems, that are not designed to document patient flows in clinical pathways. Reasons for these conditions should be better understood and studied in more depth.

Kirwin B Matthews, Maureen C Tanner

Culture influences how agile frameworks are implemented, and agility is said to be suitable in contexts where flexibility and spontaneity are emphasized. While past studies have investigated the influence of national culture on Agile implementations and practices in Western and Eastern contexts, studies focusing on a South African software development context is limited. Furthermore, few studies have focused on the effect of cultural differences within software engineering in general. The purpose of this st udy is to describe how national culture influences Agile practices within the South African software development context. In particular, the Hofstede, GLOBE and Schwartz frameworks of national culture were considered. The study was interpretive and was exe cuted using a qualitative, semi - structured interview research strategy directed at Agile practitioners in South African software development teams. The thematic analysis technique was used to analyze the data. Eleven propositions were formulated to highlight how national culture dimensions influence Agile practices. The findings reveal that various national culture dimensions influence the decision-making process, the degree of Sprint interruptions, participation in Agile ceremonies, adherence to policies and prescribed Agile practices, how teams reach agreement, approaches to process improvement and Sprint Planning . The dimensions also influence the extent to which team members are encouraged to have fun and their work-life balance, commitment to achieving Sprint Goals and the sustainable working pace , the Definition of “Done”, blocker management, how commitments are made, team cohesion, preparation for and outcomes of Sprint Retrospective, as well as the degree to which the Scrum Master is task-focused or people-focused.

Pedro Isaías

Editorial preface for Volume 17, Issue 2 of the IADIS International Journal on Computer Science and Information Systems (IJCSIS). This issue brings together peer-reviewed original research contributions covering the wide spectrum of Information Systems, Artificial Intelligence, Software Engineering, and Digital Transformation.