Die Herausforderung beim Einsatz von generativer Text-KI, wie ChatGPT, besteht darin, die Potenziale effizient zu nutzen und im Hinblick auf die Erreichung von Qualitätszielen optimal einzusetzen. Dabei ist die menschliche Eingabe in die Künstliche Intelligenz (KI) – der Prompt – entscheidend. Der vorliegende Beitrag widmet sich der Frage, wie die KI-basierte Textausgabe bei ChatGPT durch Prompt Engineering gezielt gesteuert werden kann, damit die Textqualität der generativen KI den Erfolgskriterien für Content Marketing Texte entspricht. Die Ergebnisse identifizieren eine effektive Prompt-Struktur für qualitativ hochwertige Content Marketing Texte mit ChatGPT. Insbesondere das Zero-shot Chain-of-Thought und das One-shot bzw. Few-shot Prompting erweisen sich als erfolgreich, da diese Techniken eine gezielte Steuerung des ChatGPT-Outputs in Richtung der Erfolgskriterien ermöglichen. Darüber hinaus werden die aktuellen Schwächen von KI-generierten Texten beschrieben. Dabei werden auch die Grenzen von ChatGPT deutlich, die durch eine kollaborative Wertschöpfung von Mensch und KI zur gemeinsamen Erreichung von Qualitätszielen überwunden werden können. Die theoretisch und praktisch fundierten Ergebnisse und Implikationen der Untersuchung bieten eine Orientierungshilfe für Content Marketer zur effizienten Nutzung von ChatGPT.

Tourism recommendation systems can mitigate the potential impact of choice overload on tourists. Social robots are a promising approach to provide recommendations to tourists through an engaging and intuitive user interface on sites like tourist information offices. This study investigates whether tourists perceive tourism recommendations provided via social robots as a satisfying and effective experience and whether tourists respond better to a more human or robotic design of social robot interactions. Therefore, an experiment is conducted at a real-world tourist information office where 60 tourists are exposed to either the more human or robotic version of the social robot recommender system. Their feedback is collected with a survey. The results show that the social robot is perceived positively across
all user-centric evaluation dimensions. This indicates that tourists accept social robots in real-world tourist recommendation situations and would also use them in the future.

Technology in football is increasingly used for decision making. Adoption, especially in Germany, has been slow. However, the benefits of data analytics for pre-, and post-match analysis have motivated decision makers to pay attention to the data science trend. Nowadays, football clubs from the third leagues or even amateur clubs are using technology to help them gain a competitive edge. Fan experience, both online and offline (home infront of the TV or at the stadium) is driving the next innovation stage in football. The study presented here is focused on testing and evaluation a facial recognition software on images from football coaches, just a few seconds after an important situation during the match has taken place (e.g., win, goal scored). We demonstrated that, in fact, emotion recognition software captures unexpected emotional reactions from coaches which could then be used to calculate interesting statistics and increase fan engagement and entertainment.

Digital ownership has gained attraction as a prospective domain for research and development of emerging technologies in recent years. A significant number of solutions, primarily blockchain-powered systems for digital ownership, have been developed and published aiming for widespread usage. However, the approach still appears uncommon to both digital creators and consumers community. While the majority of research in this field has been on technical aspects of implementing such solutions, there is an extreme deficiency regarding users’ viewpoints incorporated into the design and thus enlarging the barriers in mainstream adoption. This study picked the area of digital arts and shifted the focus to users’ perspectives in blockchain-based services for digital ownership in art. By adopting a qualitative approach to learn about digital creators’ behaviors and opinions, the study findings revealed various concerns about contemporary services that hinder creators’ use, their actual needs and expectations in a blockchain-based system for powering digital art products. Based on the study results, three design implications were identified to enhance the level of acceptance from the digital creator group.

Kundenfeedback im Online-Handel in Form von Produktrezensionen liefern wichtige Informationen über die Kundenwahrnehmung von Produkten. So beschreiben sie verwendete Materialien, Farben, die Passform, das Design und den Anwendungszweck eines Produkts. Das Kundenfeedback liegt hier in unstrukturierter Textform vor, weshalb zur Verarbeitung Ansätze aus dem Gebiet des Natural Language Processing und des maschinellen Lernens von Vorteil sind. In diesem Beitrag wird ein hybrider Ansatz zur Kategorisierung von Produktrezensionen vorgestellt, der die Vorteile des maschinellen Lernens des Word2Vec-Algorithmus und die der menschlichen Expertise vereint. Das daraus resultierende Datenmodell wird im Anschluss anhand einer Praxisanwendung zum Thema Produktempfehlungen demonstriert.

In e‑commerce, enormous amounts of data are generated through the interaction of customers with Web platforms. Customer feedback in the form of product reviews, for instance, is an example for unstructured data, which processing requires approaches from the fields of computer linguistics and machine learning. As an alternative to the classical approaches of supervised and unsupervised learning, which often reach their limits in the business context and the application domain of product reviews, this article presents a hybrid approach for categorizing product reviews that combines the advantages of machine learning and human expertise. The aim of this paper is to present an approach that allows to automatically extract structured topics and related aspects from product reviews based on practical requirements. Word2Vec is used to train semantic relationships between words that occur in product reviews. In this way, individual words of each review can be compared with in advance defined topic words regarding their similarity and can then be extracted from the reviews. This approach is demonstrated using around five million product reviews of the Amazon online platform. The results are getting compared with those from a common topic modelling technique.

The interdisciplinary research of neuromarketing shows that the conscious and rational consumer is only an illusion, whereas emotions have a significant influence on consumer behavior. Therefore, this study examines the effect of emotionalized e-com pages on visitors’ emotions as well as on their behavioral intention in hedonic situations. Three landing pages are conceptualized using diverse techniques of emotional boosting along with different procedures of triggering distinct levels of neuronal activity. The impact of these landing pages is examined in an online survey, generating a sample of 391 participants. The resulting dataset is analyzed by using structural equation modeling to test the proposed hypotheses. The results confirm that emotions can be triggered only by seeing a landing page of an e-com store and that these emotions influence the behavioral intentions. Additionally, the study shows a moderating effect of long-term involvement and mood and provides recommendations for appropriate and well-designed websites.

Social networks have been applied in football, or football match analysis to analyze the passing distributions between teams. However, analysis has been mostly done on a manually collected data by considering the widely adopted network metrics such as betweenness and closeness centrality. In this paper, we use professional tracking event data provided by OPTA Sports and analyze the final game of the Euro2016 between Portugal and France. We use Gephi and the NetworkX Python library and apply dynamic network analysis by integrating the timestamps of the passes. We further look into traditional performance metrics from both teams and make an attempt to connect those to the network results and the outcome of the game.

Emotions have a significant impact on the purchasing process. Due to novel affective computing approaches, affective information of users can be acquired in implicit and therefore non-intrusive manner. Recent research in the field of recommender systems indicates that the incorporation of affective user information in the prediction model has a positive impact on the recommender systems accuracy. Existing research mainly focused on product recommendations in the movie anfd music domain. Our paper investigates the impact of affective emotions on fashion products, which is one of the largest consumer industries. We integrate the users' mood and their emotion in the prediction model, and the results are compared to the baseline model using rating data only. For this, we generate a dataset with 337 participants, 64 products, and 10816 ratings. We determine the mood information using the PANAS questionnaire, and the emotion by using the SAM self-assessment method. The affective information is integrated leveraging Factorization Machines. The evaluation of the offline experiments reveals that in new item cold-start scenarios the mood information has a positive impact on the prediction accuracy, whereas the emotion information has a negative impact.

Fashion product consumer are faced with large and fast changing product o erings. e fashion purchase decision process is complex, as the consumer has to consider various in uencing factors like current fashion trends, what fashion products t to their personality, and what products t to their physical appearance like hair colors or body measures. Based on novel technologies, 3D body avatars can be reconstructed from 3D or 2D data. From these avatars, body measures can be determined. e objective of this research is to investigate the predictive performance of body measures extracted from a 3D body scanner for predicting fashion item preferences. erefore, item preferences and body scans from 200 users were collected. From the body scans, 11 body measures are extracted and integrated into a prediction model using Factorization Machines. e results from a cross-validation show, that including body measurements signi cantly improves the prediction performance of the recommendation model, especially in new user scenarios, when no information about the fashion product preferences of the active user is known.

There are many platforms on the market that support researchers and practitioners to create surveys and market studies. Nevertheless, nearly all of them focus on providing answers to textual questions. In contrast to existing systems this paper presents the concept, prototype, and evaluation of a new mobile platform for quantitative research strictly focusing on images: the SciencOmat. This platform uses pictures to evaluate products, marketing content, and other elements based on their visual attractiveness. Particular emphasis was placed on a high level of usability and user experience. The system integrates methods known from popular online dating applications (e.g., liking/disliking a product by swiping left or right) and also applies gamification elements to further drive user motivation. Next to the application and its evaluation using the User Experience Questionnaire provided by Schrepp et al. we also present the results of two exemplary image data sets.

The performance of companies depends on the ability to leverage data to create insights and to target consumers with personalized messages Like marketing content or product offerings. One key element for personalized targeting are expressive user profiles, which are the basis for predictive models to estimate individual consumers’ preferences. Traditionally user profiles are mainly based on demographic attributes like age, gender, or occupation. Due to changes in society, consumers’ behaviors are less stable, and therefore these demographic factors are less effective. Alternatively, the consumers’ lifestyle has a significant impact on their purchase and consumption behavior. This paper investigates the relationship between Facebook Likes and the lifestyle of individuals based on the activity, interests, and opinion (AIO) model. Therefore, 14482 user-Like combinations from 214 participants were collected together with lifestyle information and a correlation analysis is conducted. The results indicate weak monotonic correlations between the AIO and the Like information.