•This study proposes a guideline to create a predictive persona by using data-mining predictive algorithms.•The proposed method aims to improve the accuracy and credibility of personas.•We discuss ...the strengths, weaknesses, and using scenarios of each types of persona creating methods.
A persona is a widely used tool in the design field to help understand users’ needs, experiences, behaviors, and goals. Quantitatively creating personas is gaining importance for its greater representability, objectivity, and fast creative speed. However, quantitative personas built using traditional statistical methods may not reflect the marketing goals of decision-makers, and they mainly focus on categorizing existing users rather than predicting target customers. To address the research gap, this study proposes a guideline to create a predictive persona by using data-mining algorithms resulting in personas that can be validated. The proposed method aims to improve the accuracy and credibility of personas to effectively predict new customers for target marketing. We demonstrate a step-by-step guideline for creating the proposed predictive persona using real user data, validating its credibility, and comparing its predictive power against traditional quantitative personas. We thoroughly discuss the strengths and weaknesses of qualitative personas, traditional quantitative personas, and predictive personas while portraying suitable scenarios in which each type of persona is appropriate.
Distant pet owners suffer from negative feelings such as distress or anxiety when separating from their pet dogs. Informed by prior work on facilitating intimacy between distant couples and among ...distant family members, we present the design and evaluation of the cAMpanion ambient display to enhance distant owner's sense of connectedness with their pet dogs. cAMpanion shows three statuses of pet dogs via different light colors in real-time enabled by sensor modules installed in pet dog's living environment. We examined the impact and effectiveness of cAMpanion via a 14-day field study using mixed methods research. Our qualitative data suggested that cAMpanion helped raise awareness, while quantitative results indicated statistically significant decrease in owners' negative emotions and increase in their perceived closeness. We discussed both advantages and limitations of cAMpanion from distant owners' perspective, and provided implications for future researchers who would like to design similar display to enhance human-animal connectedness.
An apical lesion is caused by bacteria invading the tooth apex through caries. Periodontal disease is caused by plaque accumulation. Peri-endo combined lesions include both diseases and significantly ...affect dental prognosis. The lack of clear symptoms in the early stages of onset makes diagnosis challenging, and delayed treatment can lead to the spread of symptoms. Early infection detection is crucial for preventing complications. PAs used as the database were provided by Chang Gung Memorial Medical Center, Taoyuan, Taiwan, with permission from the Institutional Review Board (IRB): 02002030B0. The tooth apex image enhancement method is a new technology in PA detection. This image enhancement method is used with convolutional neural networks (CNN) to classify apical lesions, peri-endo combined lesions, and asymptomatic cases, and to compare with You Only Look Once-v8-Oriented Bounding Box (YOLOv8-OBB) disease detection results. The contributions lie in the utilization of database augmentation and adaptive histogram equalization on individual tooth images, achieving the highest comprehensive validation accuracy of 95.23% with the ConvNextv2 model. Furthermore, the CNN outperformed YOLOv8 in identifying apical lesions, achieving an F1-Score of 92.45%. For the classification of peri-endo combined lesions, CNN attained the highest F1-Score of 96.49%, whereas YOLOv8 scored 88.49%.
Music-to-visual style transfer is a challenging yet important cross-modal learning problem in the practice of creativity. Its major difference from the traditional image style transfer problem is ...that the style information is provided by music rather than images. Assuming that musical features can be properly mapped to visual contents through semantic links between the two domains, we solve the music-to-visual style transfer problem in two steps: music visualization and style transfer. The music visualization network utilizes an encoder-generator architecture with a conditional generative adversarial network to generate image-based music representations from music data. This network is integrated with an image style transfer method to accomplish the style transfer process. Experiments are conducted on WikiArt-IMSLP, a newly compiled dataset including Western music recordings and paintings listed by decades. By utilizing such a label to learn the semantic connection between paintings and music, we demonstrate that the proposed framework can generate diverse image style representations from a music piece, and these representations can unveil certain art forms of the same era. Subjective testing results also emphasize the role of the era label in improving the perceptual quality on the compatibility between music and visual content.
As the popularity of dental implants continues to grow at a rate of about 14% per year, so do the risks associated with the procedure. Complications such as sinusitis and nerve damage are not ...uncommon, and inadequate cleaning can lead to peri-implantitis around the implant, jeopardizing its stability and potentially necessitating retreatment. To address this issue, this research proposes a new system for evaluating the degree of periodontal damage around implants using Periapical film (PA). The system utilizes two Convolutional Neural Networks (CNN) models to accurately detect the location of the implant and assess the extent of damage caused by peri-implantitis. One of the CNN models is designed to determine the location of the implant in the PA with an accuracy of up to 89.31%, while the other model is responsible for assessing the degree of Peri-implantitis damage around the implant, achieving an accuracy of 90.45%. The system combines image cropping based on position information obtained from the first CNN with image enhancement techniques such as Histogram Equalization and Adaptive Histogram Equalization (AHE) to improve the visibility of the implant and gums. The result is a more accurate assessment of whether peri-implantitis has eroded to the first thread, a critical indicator of implant stability. To ensure the ethical and regulatory standards of our research, this proposal has been certified by the Institutional Review Board (IRB) under number 202102023B0C503. With no existing technology to evaluate Peri-implantitis damage around dental implants, this CNN-based system has the potential to revolutionize implant dentistry and improve patient outcomes.