Hybrid Variational Autoencoder for Recommender Systems

E-commerce platforms heavily rely on automatic personalized recommender systems, e.g., collaborative filtering models, to improve customer experience. Some hybrid models have been proposed recently to address the deficiency of existing models. However, their performances drop significantly when the dataset is sparse. Most of the recent works failed to fully address this shortcoming. At most, some of them only tried to alleviate the problem by considering either user side or item side content information. In this article, we propose a novel recommender model called Hybrid Variational Autoencoder (HVAE) to improve the performance on sparse datasets. Different from the existing approaches, we encode both user and item information into a latent space for semantic relevance measurement. In parallel, we utilize collaborative filtering to find the implicit factors of users and items, and combine their outputs to deliver a hybrid solution. In addition, we compare the performance of Gaussian distribution and multinomial distribution in learning the representations of the textual data. Our experiment results show that HVAE is able to significantly outperform state-of-the-art models with robust performance.

A Deep Content-Based Model for Persian Rumor Verification

During the development of social media, there has been a transformation in social communication. Despite their positive applications in social interactions and news spread, it also provides an ideal platform for spreading rumors. Rumors can endanger the security of society in normal or critical situations. Therefore, it is important to detect and verify the rumors in the early stage of their spreading. Many research works have focused on social attributes in the social network to solve the problem of rumor detection and verification, while less attention has been paid to content features. The social and structural features of rumors develop over time and are not available in the early stage of rumor. Therefore, this study presented a content-based model to verify the Persian rumors on Twitter and Telegram early. The proposed model demonstrates the important role of content in spreading rumors and generates a better-integrated representation for each source rumor document by fusing its semantic, pragmatic, and syntactic information. First, contextual word embeddings of the source rumor are generated by a hybrid model based on ParsBERT and parallel CapsNets. Then, pragmatic and syntactic features of the rumor are extracted and concatenated with embeddings to capture the rich information for rumor verification. Experimental results on real-world datasets demonstrated that the proposed model significantly outperforms the state-of-the-art models in the early rumor verification task. Also, it can enhance the performance of the classifier from 2% to 11% on Twitter and from 5% to 23% on Telegram. These results validate the model's effectiveness when limited content information is available.

Information on Covid-19 and Effortful Control Level: Increasing or Decreasing Anxiety?

Anxiety has increased with more information about the COVID-19 outbreak. The COVID-19 information can make people more aware and care about this pandemic. However, it also can make people feel panic and anxious. Emotional induction in the information is referring to how the emotion carried by the information can affect early adult’s anxiety, but the differences in effortful control level that each person has also define the level of anxiety someone experiences. The anxiety level related to effortful control is one dimension of temperament. This study is conducted in concern of the effect of emotional induction in COVID-19 information towards anxiety in high and low effortful control groups. This study used a quantitative method with quasi-experimental research. This study involves 218 early adults in an age range of 20-40 years old. This study used pictures from content information about COVID-19, Self-Assessment Manikin (SAM), Beck Anxiety Inventory (BAI), and Effortful Control Questionnaire (ECQ) as the measuring tools. The results of the significance test show p <0.005. This means that there is an effect of emotional induction on anxiety at both high and low effortful control levels. This study shows that emotional induction in the form of COVID-19 information reduces anxiety in the respondent group with a high level of effortful control, but increases anxiety in the respondent group with a low level of effortful control. The effect of COVID-19 information on anxiety based on effortful control in the late adult (over 65 years old) group, in this study cannot be explained.

Effect of Addition of Moringa Leaves (Moringa oleifera) on Chemical Characteristics and Nutritional Value of Chicken Sausage Chips

The leading cause of malnutrition in NTT province is the lack of balanced nutrition, one of which is macronutrients such as carbohydrates, proteins, and fats. So, there is a need for innovation from livestock products rich in readily available animal protein, one of which is chicken. Utilization of the results of heating technology Moringa leaf flour which is rich in micro and macronutrients is available in the province of NTT, so that it can provide processed products in the form of chicken sausage chips with the addition of Moringa leaf flour. This study aims to determine the effect of adding Moringa leaf flour P0, P1, P2, and P3 on chicken sausage chips' chemical characteristics and nutritional value. Each treatment P0 Moringa leaf flour (0%), P1 Moringa leaf flour (1%), P2 Moringa leaf flour (2%), P3 Moringa leaf flour (3%). The experimental method used is a simple, completely randomized design (CRD) with four treatments and four replications. Analysis of the ANOVA (Analysis of Variance) data and Duncan's further test, while the nutritional value content was calculated according to BPOM regulation NO 19 of 2019 concerning Nutrition Label Reference. The results showed that the addition of Moringa leaf flour had a significant effect on the chemical characteristics of P<0.05. According to the National Food and Drug Administration Agency (BPOM), for appropriate nutritional content, information on the nutritional value of chicken sausage chips.

Urban Collectives and insurgency to fight COVID-19: an analysis of social media content

The high incidences of COVID-19 cases in low-income communities, the context of disinformation around the disease and the vacuum of public policies have made low income communities more vulnerable to the pandemic. Considering that scenario, this paper analyzes how urban collectives have used social media to create and share narratives about COVID-19. We frame those contributions through a lens of insurgency, an area of planning studies that is based on grassroots counter-hegemonic actions. We conduct a sentiment and thematic analysis of Instagram posts of urban collectives in Fortaleza to show how social media has been appropriated as a space to cope with the virus and support insurgency. Our findings show, even though most strategies are geared towards emergency relief, collectives also harness the atmosphere of crisis brought by COVID-19 to raise awareness to other structural issues. Collectives promote insurgency by creating their own content, information and research material about COVID-19 in their communities and by partnering with institutions to scale up their claims.

Text Messages to Curb Sugar-Sweetened Beverage Consumption among Pregnant Women and Mothers: A Mobile Health Randomized Controlled Trial

Racial, ethnic, and socioeconomic disparities in childhood obesity in the United States (U.S.) originate in early life. Maternal sugar-sweetened beverage (SSB) consumption is an early life risk factor for later offspring obesity. The goal of this study was to test the effects of policy-relevant messages delivered by text messages mobile devices (mHealth) on maternal SSB consumption. In this three-arm 1-month randomized controlled trial (RCT), pregnant women or mothers of infants in predominantly Hispanic/Latino New York City neighborhoods were randomized to receive one of three text message sets: graphic beverage health warning labels, beverage sugar content information, or attention control. The main outcome was change in maternal self-reporting of average daily SSB consumption from baseline to one month. Among 262 participants, maternal SSB consumption declined over the 1-month period in all three arms. No intervention effect was detected in primary analyses. In sensitivity analyses accounting for outliers, graphic health warning labels reduced maternal SSB consumption by 28 kcal daily (95% CI: −56, −1). In this mHealth RCT among pregnant women and mothers of infants, graphic health warning labels and beverage sugar content information did not reduce maternal SSB consumption.

Characterization of different forms of kava (Piper methysticum) products by UPLC-MS/MS

There are several forms of kava (Piper methysticum) products available for human consumption, and many factors are known to influence their chemical compositions and therefore their pharmacological properties. Because of the increased popularity of kava intake, a rigorous characterization of their content diversity is prerequisite, particularly due to its known potential to cause hepatotoxicity. To understand the composition diversity of kavalactones and flavokavains in commercial kava products, we developed a UPLC-MS/MS-based analytical method for the quantification of six kavalactones (kavain, dihydrokavain, methysticin, dihydromethysticin, yangonin and desmethoxyyangonin) and two flavokavains (flavokavains A and B) and analyzed their contents in 28 different kava products in the form of capsules, tinctures, traditional aqueous suspensions and dried powders. Our results demonstrated a great variation in terms of the total and relative abundance of the analyzed kavalactones and flavokavains among the analyzed kava preparations. More importantly, the kavalactone abundance in the product label could differ up to 90% from our experimental measurements. Therefore, more rigorous and comprehensive quality control of kava products is required with respect to the content of individual kavalactones and flavokavains. Accurate content information is essential to understand the pharmacological properties and safety of different kava products.

U2-VC: one-shot voice conversion using two-level nested U-structure

Voice conversion is to transform a source speaker to the target one, while keeping the linguistic content unchanged. Recently, one-shot voice conversion gradually becomes a hot topic for its potentially wide range of applications, where it has the capability to convert the voice from any source speaker to any other target speaker even when both the source speaker and the target speaker are unseen during training. Although a great progress has been made in one-shot voice conversion, the naturalness of the converted speech remains a challenging problem. To further improve the naturalness of the converted speech, this paper proposes a two-level nested U-structure (U2-Net) voice conversion algorithm called U2-VC. The U2-Net can extract both local feature and multi-scale feature of log-mel spectrogram, which can help to learn the time-frequency structures of the source speech and the target speech. Moreover, we adopt sandwich adaptive instance normalization (SaAdaIN) in decoder for speaker identity transformation to retain more content information of the source speech while maintaining the speaker similarity between the converted speech and the target speech. Experiments on VCTK dataset show that U2-VC outperforms many SOTA approaches including AGAIN-VC and AdaIN-VC in terms of both objective and subjective measurements.

Synergy of language and engineering education: Team project as a learning technology and an assessment tool

Актуальность статьи обусловлена недостаточной изученностью вопроса интеграции языкового и инженерного образования, а также необходимостью выработки эффективных механизмов реализации компетентностного подхода средствами цифровых образовательных технологий. Цель статьи заключается в описании технологии командного проекта как формирующе-оценивающей педагогической технологии, положенной в основу организации курса «Иностранный язык» для студентов неязыковых специальностей. Авторы раскрывают методологические основания разработанной педагогической технологии, включающие принципы компетентностного, деятельностного, коммуникативного, социокультурного модульного и проектного подходов; описывают организационную, содержательную, информационно-технологическую и оценочную компоненты технологии на примере конкретного командного проекта. Доказано, что предложенная технология позволяет получить устойчивые положительные результаты в развитии предметных и надпредметных компетенций. Статья предназначена для практиков языкового образования в высшей школе. The relevance of the article is due to the insufficient study of the issue of integration of language and engineering education, as well as the need to develop effective mechanisms for implementing the competence approach by means of digital educational technologies. The purpose of the article is to describe the technology of the team project as a formative-evaluating pedagogical technology, which is the basis for the organization of the course "Foreign language" for students of non-linguistic specialties. The authors reveal the methodological foundations of the developed pedagogical technology, including the principles of competence, activity, communicative, socio-cultural modular and project approaches; describe the organizational, content, information technology and evaluation components of the technology on the example of a specific team project. It is proved that the proposed technology allows to obtain stable positive results in the development of subject and supra-subject competencies. The article is intended for practitioners of language education in higher education.

Early Math Interventions in Informal Learning Settings Coding Protocol

The purpose of document is to provide readers with the coding protocol that authors used to code experimental and quasi-experimental early mathematics intervention studies conducted in informal learning environments. The studies were conducted in homes and in museums with caregivers as intervention agents and included children between the ages of 3,0 and 8,11 years. The coding protocol includes more than 200 variables related to basic study information, participant sample size and demographics, methodological information, intervention information, mathematics content information, the control/comparison condition, outcome measures, and results and effect sizes. The coding protocol was developed for the purpose of conducting a meta-analysis; results of the meta-analysis is pending. The data set associated with this coding protocol will be available to the public at the conclusion of the grant (early 2024).