Navigation by Revealing Trade-offs for Content-Based Recommendations

Conversational recommender systems have been introduced to provide users the opportunity to give feedback on items in a turn-based dialog until a final recommendation is accepted. Tourism is a complex domain for recommender systems because of high cost of recommending a wrong item and often relatively few ratings to learn user preferences. In a scenario such as recommending a city to visit, conversational content-based recommendation may be advantageous, since users often struggle to specify their preferences without concrete examples. However, critiquing item features comes with challenges. Users might request item characteristics during recommendation that do not exist in reality, for example demanding very high item quality for a very low price. To tackle this problem, we present a novel conversational user interface which focuses on revealing the trade-offs of choosing one item over another. The recommendations are driven by a utility function that assesses the user’s preference toward item features while learning the importance of the features to the user. This enables the system to guide the recommendation through the search space faster and accurately over prolonged interaction. We evaluated the system in an online study with 600 participants and find that our proposed paradigm leads to improved perceived accuracy and fewer conversational cycles compared to unit critiquing.

Exploiting Latent Information in Recommender Systems

<p>This thesis exploits latent information in personalised recommendation, and investigates how this information can be used to improve recommender systems. The investigations span three directions: scalar rating-based collaborative filtering, distributional rating-based collaborative filtering, and distributional ratingbased hybrid filtering. In the first investigation, the thesis discovers through data analysis three problems in nearest neighbour collaborative filtering — item irrelevance, preference imbalance, and biased average — and identifies a solution: incorporating “target awareness” in the computation of user similarity and rating deviation. Two new algorithms are subsequently proposed. Quantitative experiments show that the new algorithms, especially the first one, are able to significantly improve the performance under normal situations. They do not however excel in cold-start situations due to greater demand of data. The second investigation builds upon the experimental analysis of the first investigation, and examines the use of discrete probabilistic distributional modelling throughout the recommendation process. It encompasses four ideas: 1) distributional input rating, which enables the explicit representation of noise patterns in user inputs; 2) distributional voting profile, which enables the preservation of not only shift but also spread and peaks in user’s rating habits; 3) distributional similarity, which enables the untangled and separated similarity computation of the likes and the dislikes; and 4) distributional prediction, which enables the communication of the uncertainty, granularity, and ambivalence in the recommendation results. Quantitative experiments show that this model is able to improve the effectiveness of recommendation compared to the scalar model and other published discrete probabilistic models, especially in terms of binary and list recommendation accuracy. The third investigation is based on an analysis regarding the relationship between rating, item content, item quality, and “intangibles”, and is enabled by the discrete probabilistic model proposed in the second investigation. Based on the analysis, a fundamentally different hybrid filtering structure is proposed, where the hybridisation strategy is neither linear nor sequential, but of a divide-and-conquer shape backed by probabilistic derivation. Experimental results show that it is able to outperform the standard linear and sequential hybridisation structures.</p>

Exploiting Latent Information in Recommender Systems

<p>This thesis exploits latent information in personalised recommendation, and investigates how this information can be used to improve recommender systems. The investigations span three directions: scalar rating-based collaborative filtering, distributional rating-based collaborative filtering, and distributional ratingbased hybrid filtering. In the first investigation, the thesis discovers through data analysis three problems in nearest neighbour collaborative filtering — item irrelevance, preference imbalance, and biased average — and identifies a solution: incorporating “target awareness” in the computation of user similarity and rating deviation. Two new algorithms are subsequently proposed. Quantitative experiments show that the new algorithms, especially the first one, are able to significantly improve the performance under normal situations. They do not however excel in cold-start situations due to greater demand of data. The second investigation builds upon the experimental analysis of the first investigation, and examines the use of discrete probabilistic distributional modelling throughout the recommendation process. It encompasses four ideas: 1) distributional input rating, which enables the explicit representation of noise patterns in user inputs; 2) distributional voting profile, which enables the preservation of not only shift but also spread and peaks in user’s rating habits; 3) distributional similarity, which enables the untangled and separated similarity computation of the likes and the dislikes; and 4) distributional prediction, which enables the communication of the uncertainty, granularity, and ambivalence in the recommendation results. Quantitative experiments show that this model is able to improve the effectiveness of recommendation compared to the scalar model and other published discrete probabilistic models, especially in terms of binary and list recommendation accuracy. The third investigation is based on an analysis regarding the relationship between rating, item content, item quality, and “intangibles”, and is enabled by the discrete probabilistic model proposed in the second investigation. Based on the analysis, a fundamentally different hybrid filtering structure is proposed, where the hybridisation strategy is neither linear nor sequential, but of a divide-and-conquer shape backed by probabilistic derivation. Experimental results show that it is able to outperform the standard linear and sequential hybridisation structures.</p>

A Rasch Analysis of Item Quality of the Chemical Literacy Assessment for Investigating Student’s Chemical Literacy on Chemical Rate Concepts

<p style="text-align: justify;">Assessment is a topic that continues to be developed in science education research. Assessment evaluates not only students' cognitive abilities but also their thinking skills. Therefore, in this study, an assessment that could measure students' chemical literacy was developed. Chemical literacy is a thinking skill that students must develop as part of their chemistry learning. The goal of this study was to assess item' quality, as well as student’ chemical literacy on the concept of chemical rate. The Rasch model was employed to analyze the data in this study. The results of this study depict that the developed assessment had sufficient reliability and validity to be used to assess students' chemical literacy. Furthermore, the analysis of the students’ responses to the items revealed that many students did not understand or were unaware of the context presented. These findings suggest that students' chemical literacy in the material for the reaction rate is still lacking and needs to be improved. As a result, the teacher's role in assisting students in improving their chemical literacy through chemistry learning is critical.</p>

Location of Plant Leaf Maladies Utilizing Picture Division

Agrarian efficiency is tall on which economy exceedingly depends. Typically, the as it were for cause malady discovery potted plant imperative part during horticulture park, as have to one's name illness in plants are very normal. In case legitimate care is not grip in this zone, its justification genuine impacts on potted plant through which item quality, amount or efficiency is pretentious. For occurrence a malady called small malady could be an unsafe illness establish in pine trees in Joined together condition. Location of plant infection by way of a few self-activating advantageous because it diminishes an expansive production of checking in enormous ranches of riding crop, conjointly it recognizes the side effects of illnesses they show up on plant clears out. This venture presents a calculation for image break-up strategy which is apply for framed spot and classification of plant leaf maladies. It too frames study on diverse maladies categorizes methods that can be utilized for plant leaf malady discovery. Image break-up which is an imperative part for malady discovery in plant leaf malady, is done by use inbred reckon.

Identifying Problematic Item Characteristics With Small Samples Using Mokken Scale Analysis

Researchers frequently use Mokken scale analysis (MSA), which is a nonparametric approach to item response theory, when they have relatively small samples of examinees. Researchers have provided some guidance regarding the minimum sample size for applications of MSA under various conditions. However, these studies have not focused on item-level measurement problems, such as violations of monotonicity or invariant item ordering (IIO). Moreover, these studies have focused on problems that occur for a complete sample of examinees. The current study uses a simulation study to consider the sensitivity of MSA item analysis procedures to problematic item characteristics that occur within limited ranges of the latent variable. Results generally support the use of MSA with small samples ( N around 100 examinees) as long as multiple indicators of item quality are considered.

Validation of the Quality of Dying and Death Questionnaire among the Chinese populations

Objective This study reports the evaluation of the original 31-item Quality of Dying and Death Questionnaire (QODD) using a sample of caregivers of recently deceased older adults in China, and the validation of a shortened version (QODD-C) derived from the original scale. Methods The translation was performed using a forward and back method. The full scale was tested with 212 caregivers of decedents in four regions of China. Confirmatory factor analysis tested the model fit between the full Chinese version and the original conceptual model and generated the QODD-C. The psychometric analysis was performed to evaluate the QODD-C's internal consistency, content validity, construct validity, and discriminant validity. Results A five-domain, 18-item QODD-C was identified with excellent internal consistency reliability (Cronbach's α = 0.933; split-half Pearson's value = 0.855). The QODD-C total score was significantly associated with constructs related to five domains. The caregiver's relationship with the decedent, the decedent's age at death, death reason, and death place was significantly associated with the QODD-C total score. Significance of results The QODD-C is a valid and reliable instrument for assessing the quality of dying and death among the Chinese populations.

HOTS assessment in circulatory system learning: Validity, reliability, and item quality

Biology learning, based on the curriculum-2013, requires students to be skilled at analyzing, evaluating, and creating to achieve Higher-Order Thinking Skills (HOTS). Thus, HOTS assessment instrument is crucial to be made. This study aimed to produce HOTS assessment instrument focused on circulatory system materials for XI graders. This research was conducted by using Research and Development (R&D) method with 4-D model. The subjects of this study were: two biology lecturers from Faculty of Mathematics and Science of Universitas Negeri Padang, two biology teachers, and 33 XI graders of Public Senior High School 1 Nan Sabaris. The data were collected using questionnaires which were validated by experts and analyzed using ANATES 4.09. The validity results reached as high as 87.54% (very valid) in which the empirical validity value was 88%. Meanwhile, the instrument was stated as very reliable (0.78), has a moderate difficulty, good differentiating power, and excellent option quality. Thus, as the recommendation, this instrument can be used to foster students’ HOTS.