Popular Items Research Articles

Investigating and counteracting popularity bias in group recommendations

Popularity bias is an undesirable phenomenon associated with recommendation algorithms where popular items tend to be suggested over long-tail ones, even if the latter would be of reasonable interest for individuals. Such intrinsic tendencies of the recommenders may lead to producing ranked lists, in which items are not equally covered along the popularity tail. Although some recent studies aim to detect such biases of traditional algorithms and treat their effects on recommendations, the concept of popularity bias remains elusive for group recommender systems. Therefore, in this study, we focus on investigating popularity bias from the view of group recommender systems, which aggregate individual preferences to achieve recommendations for groups of users. We analyze various state-of-the-art aggregation techniques utilized in group recommender systems regarding their bias towards popular items. To counteract possible popularity issues in group recommendations, we adapt a traditional re-ranking approach that weighs items inversely proportional to their popularity within a group. Also, we propose a novel popularity bias mitigation procedure that re-ranks items by incorporating their popularity level and estimated group ratings in two distinct strategies. The first one aims to penalize popular items during the aggregation process highly and avoids bias better, while the second one puts more emphasis on group ratings than popularity and achieves a more balanced performance regarding conflicting goals of mitigating bias and boosting accuracy. Experiments performed on four real-world benchmark datasets demonstrate that both strategies are more efficient than the adapted approach, and empowering aggregation techniques with one of these strategies significantly decreases their bias towards popular items while maintaining reasonable ranking accuracy.

RECENT TRENDS IN OSMOTIC DEHYDRATION OF FRUITS: A REVIEW

The marvel of expulsion of water from lower solvent concentration to higher concentration by a semi-perimeable membrane is osmotic dehydration, resulting in equilibrium state on both sides of the membrane. As it decreases the water activity of fruits. It is found to be most widely used application for preserving food materials. Due to colour, scent, nutritional constituents and flavour compound retention value, osmotic dehydration is favoured over other methods. Drying, shower drying, freezedrying, solidifying, vacuum packing, canning, syrup conservation (osmotic dehydration), sugar crystallization, nourishment illumination and the addition of preservatives or inert gases such as carbon dioxide are typical methods of applying these processes. The solutes used in osmotic parchedness are ordinarily sugar syrup with fruit slices or 3d shapes, and salt (sodium chloride) or vegetable brine. This can be the method of multicomponent diffusion. Water moves from fruits or vegetables to the solution in this process and certain fruit and vegetable components such as minerals, vitamins, fruit acids, etc also pass into the solution along with water to the fruits and vegetables, sugar and salt migrate. Examples that are considered are Gooseberry Murabbas (Aonla), apples, sweets, candies of Different fruits and vegetables, including pethas, parwal, Osmosis-made sweets in sugar syrup are the most popular items of commercial value available in the market made from fruits. It is processed in a brine solution before drying in pickles made from raw mango. In brine, various vegetables are handled to lower their moisture content

Exploiting Temporal Dynamics in Product Reviews for Dynamic Sentiment Prediction at the Aspect Level

Online reviews and ratings play an important role in shaping the purchase decisions of customers in e-commerce. Many researches have been done to make proper recommendations for users, by exploiting reviews, ratings, user profiles, or behaviors. However, the dynamic evolution of user preferences and item properties haven’t been fully exploited. Moreover, it lacks fine-grained studies at the aspect level. To address the above issues, we define two concepts of user maturity and item popularity, to better explore the dynamic changes for users and items. We strive to exploit fine-grained information at the aspect level and the evolution of users and items, for dynamic sentiment prediction. First, we analyze three real datasets from both the overall level and the aspect level, to discover the dynamic changes (i.e., gradual changes and sudden changes) in user aspect preferences and item aspect properties. Next, we propose a novel model of Aspect-based Sentiment Dynamic Prediction (ASDP), to dynamically capture and exploit the change patterns with uniform time intervals. We further propose the improved model ASDP+ with a bin segmentation algorithm to set the time intervals non-uniformly based on the sudden changes. Experimental results on three real-world datasets show that our work leads to significant improvements.

Enterococcal Species Associated with Slovak Raw Goat Milk, Their Safety and Susceptibility to Lantibiotics and Durancin ED26E/7

Goat milk has become a popular item of human consumption due to its originality. Enterococci are ubiquitous bacteria, and they can also be found in traditional dairy products. This study focuses on the safety of enterococci from Slovak raw goat milk and on their susceptibility to lantibiotic bacteriocins and durancin ED26E/7, which has not previously been studied. Biofilm formation ability in enterococci, virulence factor genes, enzyme production and antibiotic profile were investigated. Samples of raw goat milk (53) were collected from 283 goats in Slovakia. MALDI-TOF mass spectrometry identified three enterococcal species: Enterococcus faecium, E. hirae and E. mundtii, with dominant occurrence of the species E. faecium. Low-grade biofilm formation ability (0.1 ≤ A570 < 1.0) was found in four strains of E. faecium. Gelatinase, hyaluronidase, aggregation substance and enterococcal surface protein genes were absent in these enterococci. Gene efaAfm (adhesin) was detected in five E. faecium strains. However, it was not detected in biofilm-forming strains. Enterococci detected in Slovak raw goat milk were found not to have pathogenic potential; four strains even produced high amounts of useful β-galactosidase. The strains were susceptible to lantibiotic bacteriocin treatment and to durancin ED26E/7 as well, which represents original information in dairy production.

Personalised attraction recommendation for enhancing topic diversity and accuracy

Attraction recommendation plays an important role in tourism, such as solving information overload problems and recommending proper attractions to users. Currently, most recommendation methods are dedicated to improving the accuracy of recommendations. However, recommendation methods only focusing on accuracy tend to recommend popular items that are often purchased by users, which results in a lack of diversity and low visibility of non-popular items. Hence, many studies have suggested the importance of recommendation diversity and proposed improved methods, but there is room for improvement. First, the definition of diversity for different items requires consideration for domain characteristics. Second, the existing algorithms for improving diversity sacrifice the accuracy of recommendations. Therefore, the article utilises the topic ‘features of attractions’ to define the calculation method of recommendation diversity. We developed a two-stage optimisation model to enhance recommendation diversity while maintaining the accuracy of recommendations. In the first stage, an optimisation model considering topic diversity is proposed to increase recommendation diversity and generate candidate attractions. In the second stage, we propose a minimisation misclassification cost optimisation model to balance recommendation diversity and accuracy. To assess the performance of the proposed method, experiments are conducted with real-world travel data. The results indicate that the proposed two-stage optimisation model can significantly improve the diversity and accuracy of recommendations.

Content Pushing Over Idle Timeslots: Performance Analysis and Caching Gains

Caching holds the promise of scaling the service capability of next-generation radio access networks (RANs), thereby attracting much recent attention in the era of 6G research. To enable caching, popular content items should be proactively pushed to the user ends (UEs) in the placement phase. In practice, most content placement is only allowed to exploit idle spectrum or timeslot that is not occupied by any on-demand transmissions. In this case, however, the performance analysis and optimizations of practical caching gains remain open. In this paper, we are interested in how pushing, as a secondary service, improves the overall performance in terms of energy efficiency and latency reduction. To this end, we formulate a Markovian queueing model, the caching gains of which are optimized via linear programming (LP) with acceptable computational complexity. Our work demonstrates that the peak traffic load due to burst on-demand transmissions, as primary services, can be effectively offloaded by pushing over idle timeslots, leading to substantially reduced power consumption and queueing delay.

Classification and Evaluation for Microblog Popularity Prediction

In recent years, with the rapid development of the Internet, especially the mobile Internet, social networks have entered the stage of vigorous development and become one of the main sources of information. User-generated contents (UGC) on social platforms can spread information along social networks at an astonishing speed. Existing literature has proposed many prediction methods for the popularity prediction on social networks. This paper presents a classification and establishes a unified evaluation framework of popularity prediction methods for microblogs. More specifically, we divide these mainstream prediction methods into four types: feature based methods, time series methods, collaborative filtering methods and deep learning methods and conduct experiments on the real-world weibo data using these methods to predict. Finally, according to four indicators, including accuracy, efficiency, robustness and bias, we evaluate and compare the methods. Based on the prediction and evaluation results, this paper summarizes and draws the following research conclusions:(1) The deep learning method has the characteristics of high accuracy, high robustness and low bias. The DeepFM method, one of the deep learning methods, performs better than the other three prediction methods when using temporal data as its input. (2) The feature based methods only using temporal features are basically consistent with those using all available features, indicating that the temporal feature has strong prediction power. Therefore, the ‘peeking’ strategy that monitors the early response of users in the initial period after the items are posted is effective. Additionally, the predictive power of temporary features can be further amplified in time series methods and deep learning methods. (3) Due to the sparse user-item interaction in social networks, the accuracy and efficiency of collaborative filtering methods are low, which makes it impossible to predict the popularity of items in social networks well.

Targeted LC-MS/MS combined with multilocus DNA metabarcoding as a combinatory approach to determine the amount and the source of pyrrolizidine alkaloids contamination in popular cooking herbs, seeds, spices and leafy vegetables

ABSTRACT Pyrrolizidine alkaloids (PAs) and their corresponding N-oxides (PANOs) are natural protoxins biosynthesised by many plant species and are responsible for occasional fatal intoxication outbreaks due to the consumption of contaminated food. However, only in rare cases has the origin of the contamination been determined. Although their presence has been studied in many matrices, occurrence data on popular widely used cooking herbs, seeds, spices and leafy vegetables is very scarce. Therefore, a systematic study on the occurrence of PA/PANOs contaminations in these popular herbal items, available on the Belgian market, was performed, by means of a validatedtargeted LC-MS/MS analysis, followed by multilocus DNA metabarcoding to track back the origin of the contamination for seven highly to moderately contaminated samples. Our results clearly indicate that 21% of the seed spices and 25% seed-based aromatic mixes contain an amount higher than 400 μg of the 30 summed targeted PAs and PANOs per kg, the value which is currently under discussion by the European member states to be set as a maximum threshold. For both the herbs and the herb-based mixes only 7% of analysed samples exceeded these levels. As a proof of concept, multilocus DNA metabarcoding was performed on six highly contaminated samples, belonging to each subtype matrix, containing high levels of heliotrine type of PA/PANOs. Each time the analysis demonstrated the presence of DNA from a plant species belonging to the plant genus Heliotropium. Moreover, a contaminated leafy vegetable sample, containing solely senecionine type PA/PANOs, contained DNA from Senecio vulgaris. Taken together, it can be stated that the proposed combinatory chemical and molecular techniques could be used to verify if a PA/PANO(s) contamination occurred in these popular cooking items and to pinpoint the origin the contamination, which is pivotal in the case of a detrimental intoxication or intoxication outbreak.

AoI-Delay Tradeoff in Mobile Edge Caching With Freshness-Aware Content Refreshing

Mobile edge caching can effectively reduce service delay but may introduce information staleness, calling for timely content refreshing. However, content refreshing consumes additional transmission resources and may degrade the delay performance of mobile systems. In this work, we propose a freshness-aware refreshing scheme to balance the service delay and content freshness measured by Age of Information (AoI). Specifically, the cached content items will be refreshed to the up-to-date version upon user requests if the AoI exceeds a certain threshold (named as refreshing window). The average AoI and service delay are derived in closed forms approximately, which reveals an AoI-delay tradeoff relationship with respect to the refreshing window. In addition, the refreshing window is optimized to minimize the average delay while meeting the AoI requirements, and the results indicate to set a smaller refreshing window for the popular content items. Extensive simulations are conducted on the OMNeT++ platform to validate the analytical results. The results indicate that the proposed scheme can restrain frequent refreshing as the request arrival rate increases, whereby the average delay can be reduced by around 80% while maintaining the AoI below one second in heavily-loaded scenarios.

3D Ziyarat: Lenticularity and technologies of the moving image in material and visual piety

Images in movement and images of movement take a central place in iterations of Islamic material and visual piety. Yet the still image remains the default object of inquiry in most studies of devotional visuality. Taking movement rather than stasis as the prerequisite for engagement, many popular decorative items and technologies that share the status of “moving images” evoke the affective experience of pilgrimage. In what follows, close analysis of a repertoire of one such image technology, the lenticular print, formed through its circulation in Pakistan, provides insights into the kind of engagement expected from images whose efficacy is governed by their mobility. By testing the application of some significant theories on the phenomenology of film experience, combined with ways of understanding the forms of presence elicited by virtual pilgrimage, this essay attempts to better grasp the kinds of engagement invited by devotional objects characterized by their evocation of movement.

P0675 / #906: PEDIATRIC CRITICAL CARE FELLOW ENGAGEMENT WITH AN OPEN-ACCESS E-LEARNING RESOURCE

Aims & Objectives: Pediatric critical care medicine (PCCM) fellows must provide acute, complex clinical care while addressing numerous knowledge gaps, and so may particularly benefit from online educational support. Limited data illustrate the e-learning behaviors or interests of PCCM fellows specifically. Such information may help guide design and targeting of educational interventions. To explore this, we characterized PCCM fellow usage of OPENPediatrics, a worldwide open-access e-learning resource, from June 2018 through June 2019. Methods: IRB-approved retrospective analysis of automatically collected activity log data from users who self-identified as PCCM fellows. The most popular resources (based on percent of fellows who accessed the resource) are reported. Timestamps were used to calculate the duration of time spent by individual users on the website. Results: 299 fellows from 39 countries were included. The distribution of time spent on the site was bimodal, with 105/299 (35%) of users logging <=60 minutes on the site in the analyzed interval, and 100/299 (33%) logging >=240 minutes (Figure 1). These users accessed an average of 9 resources on the site (range 1-89). Table 1 reports the most frequently accessed resources by this cohort of users.Conclusions: The majority of PCCM fellows spent at least 30 minutes on the site. Over 33% of included fellows spent >=4 hours using the site in one year. Courses and simulators were especially popular. Specific popular items included PCCM fellowship-specific curricula, materials for cardiology and congenital heart disease, and interactive online simulators teaching mechanical ventilation and hemodialysis. Our findings can help shape educational interventions and strategies for PCCM fellows.

Socially Desirable Responding in Experience Sampling

Abstract. Experience sampling often makes use of items that are similar to personality questionnaire items. Arguably, this opens up for item-popularity effects, where some respondents react to the items’ level of evaluative phrasing, causing a separate factor. Gauging the risk of item popularity effects in experience sampling is important since the multifactorial aspect of the responses to the items may cause spurious correlations. We investigate this in one original study and two existing datasets. The results reveal that evaluativeness in experience sampling items creates the same type of problems as in self-rating inventories. We conclude that personality researchers need to be aware that the experience sampling method is not vaccinated against socially desirable responding, and that careful phrasing of items promotes purer personality measures. This allows for more optimal testing of theoretical models of personality, as the fit between data and model will not concern variance in socially desirable responding but in the relevant constructs.

Demystifying the Dark Web Opioid Trade: Content Analysis on Anonymous Market Listings and Forum Posts.

BackgroundOpioid use disorder presents a public health issue afflicting millions across the globe. There is a pressing need to understand the opioid supply chain to gain new insights into the mitigation of opioid use and effectively combat the opioid crisis. The role of anonymous online marketplaces and forums that resemble eBay or Amazon, where anyone can post, browse, and purchase opioid commodities, has become increasingly important in opioid trading. Therefore, a greater understanding of anonymous markets and forums may enable public health officials and other stakeholders to comprehend the scope of the crisis. However, to the best of our knowledge, no large-scale study, which may cross multiple anonymous marketplaces and is cross-sectional, has been conducted to profile the opioid supply chain and unveil characteristics of opioid suppliers, commodities, and transactions.ObjectiveWe aimed to profile the opioid supply chain in anonymous markets and forums via a large-scale, longitudinal measurement study on anonymous market listings and posts. Toward this, we propose a series of techniques to collect data; identify opioid jargon terms used in the anonymous marketplaces and forums; and profile the opioid commodities, suppliers, and transactions.MethodsWe first conducted a whole-site crawl of anonymous online marketplaces and forums to solicit data. We then developed a suite of opioid domain–specific text mining techniques (eg, opioid jargon detection and opioid trading information retrieval) to recognize information relevant to opioid trading activities (eg, commodities, price, shipping information, and suppliers). Subsequently, we conducted a comprehensive, large-scale, longitudinal study to demystify opioid trading activities in anonymous markets and forums.ResultsA total of 248,359 listings from 10 anonymous online marketplaces and 1,138,961 traces (ie, threads of posts) from 6 underground forums were collected. Among them, we identified 28,106 opioid product listings and 13,508 opioid-related promotional and review forum traces from 5147 unique opioid suppliers’ IDs and 2778 unique opioid buyers’ IDs. Our study characterized opioid suppliers (eg, activeness and cross-market activities), commodities (eg, popular items and their evolution), and transactions (eg, origins and shipping destination) in anonymous marketplaces and forums, which enabled a greater understanding of the underground trading activities involved in international opioid supply and demand.ConclusionsThe results provide insight into opioid trading in the anonymous markets and forums and may prove an effective mitigation data point for illuminating the opioid supply chain.

PPNW: personalized pairwise novelty loss weighting for novel recommendation

Most works of recommender systems focus on providing users with highly accurate item predictions based on the assumption that accurate suggestions can best satisfy users. However, accuracy-focused models also create great system bias towards popular items and, as a result, unpopular items rarely get recommended and will stay as “cold items” forever. Both users and item providers will suffer in such scenario. To promote item novelty, which plays a crucial role in system robustness and diversity, previous studies focus mainly on re-ranking a top-N list generated by an accuracy-focused base model. The re-ranking algorithm is thus completely independent of the base model. Eventually, these frameworks are essentially limited by the base model and the separated 2 stages cause greater complication and inefficiency in providing novel suggestions. In this work, we propose a personalized pairwise novelty weighting framework for BPR loss function, which covers the limitations of BPR and effectively improves novelty with negligible decrease in accuracy. Base model will be guided by the novelty-aware loss weights to learn user preference and to generate novel top-N list in only 1 stage. Comprehensive experiments on 3 public datasets show that our approach effectively promotes novelty with almost no decrease in accuracy.

Quality Assessment of Ethnic Fermented Product Nga-Pi Traditionally Produced at Cox’s Bazar Region

Nga-pi, an ethnic fermented product traditionally produced at Cox’s Bazar region by Rakhine people is one of the popular food items of that area. To assess the quality of raw shrimp used for Nga-pi production and traditionally produced Nga-pi were collected. Some amount of the collected Nga-pi sample (from producers of both areas) were stored at refrigeration temperature (5 to 8°C) for 90 days to observe the changes in nutritive values, TVB-N and SPC during the storage. The results of proximate composition analysis for the samples showed that-the moisture content ranged from 58.29% to 61.06%, the crude protein content ranged from 25.88% to 27,12%, the crude lipid content ranged from 4.18% to 5.12% and ash content ranged from 7.30% to 8.40%. On the other hand the range for TVB-N value was found from 25.06 mg/100g to 34.02 mg/100g and SPC value from 1.69×105 CFU/g to 4.27x 106in Nga-pi samples. The result of the Nga-pi sample stored in air-tight polythene pack at refrigeration temperature (5 to 8°C) for 90 days showed that- the percent moisture, percent ash content, TVB-N value and SPC value increased but the percent protein and lipid content decreased with the progress of storage time.&#x0D; Environ. Sci. &amp; Natural Resources, 12(1&amp;2): 83-90, 2019

Evaluation of microbial contamination level and the drug susceptible pattern of the isolates cultivated from famous dessert food

Present study endeavored to evaluate the microbial contamination level along with their drug resistant pattern in some popular desert food items collected from different food shops in Dhaka city, Bangladesh. The microbial evaluation was conducted through conventional cultural methods and drug susceptibility test was executed through disc diffusion method. All the samples were found to be contaminated with heterotrophic bacteria as well as fungi within the range of 103 to 105 CFU/g. In case of specific microflora, the growth of Staphylococcus spp. was very high in sweet, faluda, milk cake and ice cream as estimated up to 105 CFU/g, halua and sweet yogurt showed 104 CFU/g while rest of the samples revealed 103 CFU/g. E. coli was found only in faluda and ice cream up to 103 CFU/g whereas Klebsiella spp. was estimated in all the samples within the range of 102 CFU/g to 105 CFU/g. Salmonella spp., Pseudomonas spp. and Bacillus spp. were totally absent in all the samples. Most of the isolates were found to be resistant against most of the antibiotics. Meanwhile, streptomycin (10 µg), gentamicin (10 µg), azithromycin (15 µg), and nalidixic acid (5 µg) were effective drug against both E. coli and Staphylococcus spp.

Item trend learning for sequential recommendation system using gated graph neural network

Recommendation system, or recommender system, is widely used in online Web applications like e-commerce Web sites and movie review Web sites. Sequential recommender put more emphasis upon user’s short-term preference through exploiting information from its recent history. By incorporating the user short-term preference into the recommendation, the algorithm achieves a higher accuracy, which proves that a more accurate user portrait or representation boosts the performance to a great extent. Intuitionally, we seek to improve the current item representation modeling via incorporating the item trend information. Most of the recommendation models neglect the importance of the ever-changing item popularity. To this end, this paper introduces a novel sequential recommendation approach dubbed TRec. TRec learns the item trend information from the implicit user interaction history and incorporates the item trend information into the subsequent item recommendation tasks. After that, a self-attention mechanism is used for better representation. We also investigate alternative ways to model the proposed item trend representation; we evaluate two variant models which leverage the power of gated graph neural network upon the item trend representation modeling to boost the representation ability. We conduct extensive experiments with seven baseline methods on four benchmark datasets. The empirical results show that our proposed approach outperforms the state-of-the-art models as high as 18.2%. The experiment result displays the effectiveness in item trend information learning while with low computational complexity as well. Our study demonstrates the importance of item trend information in recommendation system.

Byproduct utilization of watermelon to develop watermelon rind flour based cookies

Bakery products are popular among different cross sections of population in India and many bakery products are nutritional and have become important part of diet of the people. Cookies are most popular snack items that are small, usually round and flat made from stiff, sweetened dough, and baked. In the present study, Cookies were developed by incorporating watermelon rind flour at different levels like 10, 20, and 30 per cent with refined wheat flour and compared with control (100% refined wheat flour). The sensory scores revealed that the overall acceptability of cookies with 30% incorporation had good acceptable score (8.23) whereas for control cookies sensory score was (8.04). The nutrient composition of best accepted cookies had protein (4.68g), fat (30.04g), crude fibre (3.25g) and carbohydrate (45.69g). Also, cookies had good amount of calcium, iron and phosphorous i.e. 43.11mg, 2.30mg, and 71.81mg respectively per 100g of cookies. Hence watermelon rind flour which is nutritious by-product of watermelon can be utilized in bakery products to upscale the nutritional profile.

Communication-Based Book Recommendation in Computational Social Systems

This paper considers current personalized recommendation approaches based on computational social systems and then discusses their advantages and application environments. The most widely used recommendation algorithm, personalized advice based on collaborative filtering, is selected as the primary research focus. Some improvements in its application performance are analyzed. First, for the calculation of user similarity, the introduction of computational social system attributes can help to determine users’ neighbors more accurately. Second, computational social system strategies can be adopted to penalize popular items. Third, the network community, identity, and trust can be combined as there is a close relationship. Therefore, this paper proposes a new method that uses a computational social system, including a trust model based on community relationships, to improve the user similarity calculation accuracy to enhance personalized recommendation. Finally, the improved algorithm in this paper is tested on the online reading website dataset. The experimental results show that the enhanced collaborative filtering algorithm performs better than the traditional algorithm.

Cluster-based quotas for fairness improvements in music recommendation systems

Increasingly, music recommendations are influencing user listening behavior, which naturally impacts the music industry, as well as the cultural and social aspects of our society. This has opened up a research area, namely the identification of biases introduced by recommender systems in the music context. Recent research, which focused on users of the Last.fm platform found that state-of-the-art music recommendation systems frequently tend to favor already popular items. To this end, we propose a new method for predicting music recommendations, which relies on a cluster-based quotas system. By assuming the distribution of the popularity of artists to have a latent variable that is estimated with a Gaussian Mixture, we find the underlying popularity clusters and use them to make the predictions by quotas that relate to each cluster mixing proportion, in a way that the resulting popularity distribution in the recommendations is closer to the popularity distribution seen in the data. In our experiments with the Last.fm data, our final predictions increased the recommendation frequencies of less popular artists, while preserving the specific characteristics of each algorithm. The GAP(g) $$_r$$ and mean average error (MAE) metrics are improved, showing that our recommendations are more accurate and approximate to the expected popularity of the artists in each user profile.