A collaborative filtering algorithm based on item labels and Hellinger distance for sparse data

The Neighbourhood-based collaborative filtering (CF) algorithm has been widely used in recommender systems. To enhance the adaptability to the sparse data, a CF with new similarity measure and prediction method is proposed. The new similarity measure is designed based on the Hellinger distance of item labels, which overcomes the problem of depending on common-rated items (co-rated items). In the proposed prediction method, we present a new strategy to solve the problem that the neighbour users do not rate the target item, that is, the most similar item rated by the neighbour user is used to replace the target item. The proposed prediction method can significantly improve the utilisation of neighbours and obviously increase the accuracy of prediction. The experimental results on two benchmark datasets both confirm that the proposed algorithm can effectively alleviate the sparse data problem and improve the recommendation results.

Development of Candidate Teachers’ Problem Solving Ability With the Audience Response System

In this study, the Audience Response System was investigated as an additional tool for interaction, and its effects on the educational environment were examined. The system was implemented at the Faculty of Sports Sciences of Trakya University in the fall semester of the 2019-2020 academic year. A pre-test of 20 questions, which was asked in the educational sciences section of the public personnel selection examination and had a similar item difficulty index, was applied to the experimental and control groups prior to the implementation of the ARS. Then, the experimental group was asked to solve the educational sciences questions with the help of the ARS-supported lectures, which were delivered 4 h a week for a total of 16 h. The same implementation was imposed on the control group without the ARS support and with the classical recitation method. A post-test of 20 questions with a similar item difficulty index was administered to both groups after this test. Data were analyzed using the SPSS 25.0 package program. A t-test was used to determine the differences between the arithmetic mean of the pre-test and post-test scores of the students. Because the unequaled control group method was used in the experiment design, the “ANOVA for Repeated Measurements” was used for intragroup and intergroup comparisons. In conclusion, it was determined that the implementation of interactive interaction technologies in the educational environment will capture the interest of students and amplify their motivation levels. The results of the study support the conclusion that the ARS system stimulates the sensory organs in terms of understanding the subject, thereby increasing the level of learning.

A Psychometric Evaluation of the Short Grit Scale

Grit, the passion and perseverance for long-term goals, has received attention from personality psychologists because it predicts success and academic achievement. Grit has also been criticized as simply another measure of self-control or conscientiousness. A precise psychometric representation of grit is needed to understand how the construct is unique and how it overlaps with other constructs. Previous research suggests that the Short Grit Scale (Grit-S) has several psychometric limitations, such as uncertain factor structure within and across populations, uncertainty about reporting total or subscale scores, and different assessment precision at low and high levels on the construct. We conducted modern psychometric techniques including parallel analysis, measurement invariance, extrinsic convergent validity, and Item Response Theory models on two American samples. Our results suggest that the Grit-S is essentially unidimensional and that there is construct overlap with the self-control construct. Subscale factors were the result of an item doublet, where two items had highly correlated uniquenesses, showed similar item information, and were more likely to exhibit measurement bias. Findings replicated across samples. Finally, we discuss recommendations for the use of the Grit-S based on the theoretical interpretation of the unidimensional factor and our empirical findings.

Product Lineups: The More You Search, The Less You Find

Consumers often try to visually identify a previously encountered product among a sequence of similar items, guided only by their memory and a few general search terms. What determines their success at correctly identifying the target product in such “product lineups”? The current research finds that the longer consumers search sequentially, the more conservative and—ironically—inaccurate judges they become. Consequently, the more consumers search, the more likely they are to erroneously reject the correct target when it finally appears in the lineup. This happens because each time consumers evaluate a similar item in the lineup, and determine that it is not the option for which they have been looking, they draw an implicit inference that the correct target should feel more familiar than the similar items rejected up to that point. This causes the subjective feeling of familiarity consumers expect to experience with the true target to progressively escalate, making them more conservative but also less accurate judges. The findings have practical implications for consumers and marketers, and make theoretical contributions to research on inference-making, online search, and product recognition.

Effects of Advocacy Banners after Abandoning Products in Online Shopping Carts

Aim/Purpose: This study empirically analyzed and examined the effectiveness of the online advocacy banners on customers’ reactions to make replacements with the similar products in their shopping carts. Background: When a product in a shopping cart is removed, it might be put back into the cart again during the same purchase or it may be bought in the future. Otherwise, it might be abandoned and replaced with a similar item based on the customer’s enquiry list or on the recommendation of banners. There is a lack of understanding of this phenomenon in the existing literature, pointing to the need for this study. Methodology: With a database from a Taiwanese e-retailer, data were the tracks of empirical webpage clickstreams. The used data for analyses were particularly that the products were purchased again or replaced with the similar ones upon the advocacy banners being shown when they were removed from customers’ shopping carts. Few pre-defined Apriori rules as well as similarity algorithm, Jaccard index, were applied to derive the effectiveness. Contribution: This study addressed a measurement challenge by leveraging the information from clickstream data – particularly clickstream data behavior. These data are most useful to observe the real-time behavior of consumers on websites and also are applied to studying click-through behavior, but not click-through rates, for web banners. The study develops a new methodology to aid advertisers in evaluating the effectiveness of their banner campaign. Findings: The recommending/advocating titles of “you probably are interested” and “the most viewed” are not significantly effective on saving back customers’ removed products or repurchasing similar items. For the banners entitled “most buy”, “the most viewed” might only show popularity of the items, but is not enough to convince them to buy. At the current stage on the host website, customers may either not trust in the host e-retailer or in such mechanism. Additionally, the advocating/recommending banners only are effective on the same customer visits and their effects fade over time. As time passes, customers’ impressions of these banners may become vague. Recommendations for Practitioners: One managerial implication is more effective adoption of advocacy/recommendation banners on e-retailing websites. Another managerial implication is the evaluation of the advocacy/recommendation banners. By using a data mining technique to find the association between removed products and restored ones in e-shoppers’ shopping carts, the approach and findings of this study, which are important for e-retailing marketers, reflect the connection between the usage of banners and the personalized purchase changes in an individual customer’s shopping cart. Recommendation for Researchers: This study addressed a new measurement which challenges to leverage the information from clickstream data instead of click-through rates – particularly retailing webpages browsing behavior. These data are most useful to observe the real-time behavior of consumers on websites and also are applied to studying click-through behavior. Impact on Society: Personalization has become an important technique that allows businesses to improve both sales and service relationships with their online customers. This personalization gives e-marketers the ability to deliver real effectiveness in the use of banners. Future Research: The effectiveness is time- and case-sensible. Business practitioners and academic researchers are encouraged to apply the mining methodology to longevity studies, specific marketing campaigns of advertising and personal recommendations, and any further recommendation algorithms.

The Hippocampus Generalizes across Memories that Share Item and Context Information

Episodic memory is known to rely on the hippocampus, but how the hippocampus organizes different episodes to permit their subsequent retrieval remains controversial. One major area of debate hinges on a discrepancy between two hypothesized roles of the hippocampus: differentiating between similar events to reduce interference and assigning similar representations to events that share overlapping items and contextual information. Here, we used multivariate analyses of activity patterns measured with fMRI to characterize how the hippocampus distinguishes between memories based on similarity at the level of items and/or context. Hippocampal activity patterns discriminated between events that shared either item or context information but generalized across events that shared similar item–context associations. The current findings provide evidence that, whereas the hippocampus can reduce mnemonic interference by separating events that generalize along a single attribute dimension, overlapping hippocampal codes may support memory for events with overlapping item–context relations. This lends new insights into the way the hippocampus may balance multiple mnemonic operations in adaptively guiding behavior.