DISA

• When? June 2nd 14.00 – 16.00
• Where? D1172 – Växjö (link will provided for those who wants to attend online)
• Registration: We would like to know how many that will attend onsite/online in order to get some fika for those onsite. So please register by May 30^th  https://forms.gle/QHnUTeGkyYdb8dDe9 

Agenda

14.00-14.10     Welcome and practical information from Welf Löwe

14.10-14.55     Presentation and discussion: Title: Towards Better Product Quality: Identifying Legitimate Quality Issues through NLP & Machine Learning Techniques – Rakshanda Jabeen, Industry PhD-student at Electrolux Professional

14.55 – 15.05  Coffee break

15.05 – 15.50  Presentation and discussion – Title: Clustering and Modeling Large Social Networks – Masoud Fatemi, DISA PhD-student from Digital Humanities

15.50 -16.00    Sum up and plan for our next seminar on September 1^st

Abstracts

Title: Towards Better Product Quality: Identifying Legitimate Quality Issues through NLP & Machine Learning Techniques – Rakshanda Jabeen, Industry PhD-student at Electrolux Professional

Manufacturers of high-end professional products are committed to delivering outstanding customer-quality experiences. They maintain databases of customer complaints and repair service jobs data to monitor product quality. Analysing the text data from service jobs can help identify common problems, recurring issues, and patterns that impact customer satisfaction, and aid manufacturers in taking corrective actions to improve product design, manufacturing processes, and customer support services. However, distinguishing legitimate quality issues from a brief, domain-specific text in service jobs remains a challenge. This study aims to automate the classification of technical service repair job data into legitimate quality issues or non-issues to assist individuals in the quality field department in a large company. To achieve this goal, we developed a comprehensive pipeline based on natural language processing and machine learning techniques including raw text pre-processing, dealing with imbalance class distribution, feature extraction, and classification. In this study, we evaluate several feature extraction and machine learning classification methods and perform the Friedman test followed by Nemenyi post-hoc analysis to find the best-performing model. Our results show that the passive-aggressive classifier achieved the highest average accuracy of 94%, and 89% average macro F1- score when trained on TF-IDF.

Title: Clustering and Modeling Large Social Networks – Masoud Fatemi, DISA PhD-student from Digital Humanities

Looking for a digital reflection of what humans do and how they act? Consider social net­works (SN); we connect and share. Over the last decade, social media platforms such as Twitter and Facebook have been extensively utilized in multidisciplinary research, particularly in the fields of computer science and linguistics. These platforms provide access to vast amounts of data, thereby significantly improving empirical accuracy and often presenting fresh perspectives on previously explored topics.

In my presentation I will employ two distinct viewpoints to examine Twitter data from the Nordic region. To approach this task from a computer science perspective, the analysis will involve clustering Nordic Twitter users according to their connections with other users. This will entail creating an undirected graph based on mutual following relationships, clustering the resulting graph into five distinct clusters utilizing a recent M-algorithm, and comparing the cluster results to the users’ geographical locations. The research findings demonstrate that users tend to cluster strongly based on their home country, with minimal cross-border interaction, despite the region’s physical borderless attributes in daily life, cultural and linguistic similarities, and the fact that four of the countries share a common Scandinavian language.

Approaching the issue through the lens of sociolinguistics, I plan to reassess the social network model of linguistic change. The model proposes that networks are characterized by ties of varying strength that impact the dissemination of novel information, with weak ties promoting change and strong ties reinforcing norms. However, the model is better suited for studying small ego networks and requires testing for its predictive capacity in large digital networks of mobile individuals. In my presentation, I will explore the relationship between the innovation rate and various types of connections within a large-scale network, emphasizing the role of network size as a crucial component of the theory. The results suggest that network size influences the role of ties, and that the distinction between weak and slightly stronger ties diminishes when the network size exceeds approximately 120 nodes. This finding aligns with previous research in social network studies and highlights the need for further exploration of computational sociolinguistics.