Dynamic Graph Reasoning for Conversational Open-Domain Question Answering

In recent years, conversational agents have provided a natural and convenient access to useful information in people’s daily life, along with a broad and new research topic, conversational question answering (QA). On the shoulders of conversational QA, we study the conversational open-domain QA problem, where users’ information needs are presented in a conversation and exact answers are required to extract from the Web. Despite its significance and value, building an effective conversational open-domain QA system is non-trivial due to the following challenges: (1) precisely understand conversational questions based on the conversation context; (2) extract exact answers by capturing the answer dependency and transition flow in a conversation; and (3) deeply integrate question understanding and answer extraction. To address the aforementioned issues, we propose an end-to-end Dynamic Graph Reasoning approach to Conversational open-domain QA (DGRCoQA for short). DGRCoQA comprises three components, i.e., a dynamic question interpreter (DQI), a graph reasoning enhanced retriever (GRR), and a typical Reader, where the first one is developed to understand and formulate conversational questions while the other two are responsible to extract an exact answer from the Web. In particular, DQI understands conversational questions by utilizing the QA context, sourcing from predicted answers returned by the Reader, to dynamically attend to the most relevant information in the conversation context. Afterwards, GRR attempts to capture the answer flow and select the most possible passage that contains the answer by reasoning answer paths over a dynamically constructed context graph . Finally, the Reader, a reading comprehension model, predicts a text span from the selected passage as the answer. DGRCoQA demonstrates its strength in the extensive experiments conducted on a benchmark dataset. It significantly outperforms the existing methods and achieves the state-of-the-art performance.

Dual-View Conditional Variational Auto-Encoder for Emotional Dialogue Generation

Emotional dialogue generation aims to generate appropriate responses that are content relevant with the query and emotion consistent with the given emotion tag. Previous work mainly focuses on incorporating emotion information into the sequence to sequence or conditional variational auto-encoder (CVAE) models, and they usually utilize the given emotion tag as a conditional feature to influence the response generation process. However, emotion tag as a feature cannot well guarantee the emotion consistency between the response and the given emotion tag. In this article, we propose a novel Dual-View CVAE model to explicitly model the content relevance and emotion consistency jointly. These two views gather the emotional information and the content-relevant information from the latent distribution of responses, respectively. We jointly model the dual-view via VAE to get richer and complementary information. Extensive experiments on both English and Chinese emotion dialogue datasets demonstrate the effectiveness of our proposed Dual-View CVAE model, which significantly outperforms the strong baseline models in both aspects of content relevance and emotion consistency.

Legal Judgment Elements Extraction Approach with Law Article-aware Mechanism

Legal judgment elements extraction (LJEE) aims to identify the different judgment features from the fact description in legal documents automatically, which helps to improve the accuracy and interpretability of the judgment results. In real court rulings, judges usually need to scan both the fact descriptions and the law articles repeatedly to find out the relevant information, and it is hard to acquire the key judgment features quickly, so legal judgment elements extraction is a crucial and challenging task for legal judgment prediction. However, most existing methods follow the text classification framework, which fails to model the attentive relations of the law articles and the legal judgment elements. To address this issue, we simulate the working process of human judges, and propose a legal judgment elements extraction method with a law article-aware mechanism, which captures the complex semantic correlations of the law article and the legal judgment elements. Experimental results show that our proposed method achieves significant improvements than other state-of-the-art baselines on the element recognition task dataset. Compared with the BERT-CNN model, the proposed “All labels Law Articles Embedding Model (ALEM)” improves the accuracy, recall, and F1 value by 0.5, 1.4 and 1.0, respectively.

I Know What You Need: Investigating Document Retrieval Effectiveness with Partial Session Contexts

Reducing user effort in finding relevant information is one of the key objectives of search systems. Existing approaches have been shown to effectively exploit the context from the current search session of users for automatically suggesting queries to reduce their search efforts. However, these approaches do not accomplish the end goal of a search system—that of retrieving a set of potentially relevant documents for the evolving information need during a search session. This article takes the problem of query prediction one step further by investigating the problem of contextual recommendation within a search session. More specifically, given the partial context information of a session in the form of a small number of queries, we investigate how a search system can effectively predict the documents that a user would have been presented with had he continued the search session by submitting subsequent queries. To address the problem, we propose a model of contextual recommendation that seeks to capture the underlying semantics of information need transitions of a current user’s search context. This model leverages information from a number of past interactions of other users with similar interactions from an existing search log. To identify similar interactions, as a novel contribution, we propose an embedding approach that jointly learns representations of both individual query terms and also those of queries (in their entirety) from a search log data by leveraging session-level containment relationships. Our experiments conducted on a large query log, namely the AOL, demonstrate that using a joint embedding of queries and their terms within our proposed framework of document retrieval outperforms a number of text-only and sequence modeling based baselines.

Sentiment Analysis in Hindi—A Survey on the State-of-the-art Techniques

Sentiment Analysis (SA) has been a core interest in the field of text mining research, dealing with computational processing of sentiments, views, and subjective nature of the text. Due to the availability of extensive web-based data in Indian languages such as Hindi, Marathi, Kannada, Tamil, and so on. It has become extremely significant to analyze this data and recover valuable and relevant information. Hindi being the first language of the majority of the population in India, SA in Hindi has turned out to be a critical task particularly for companies and government organizations. This research portrays a systematic review specifically in the field of Hindi SA. The major contribution of this article includes the categorization of numerous articles based on techniques that have attracted researchers in performing SA tasks in Hindi language. This survey classifies these state-of-the-art computational intelligence techniques into four major categories namely lexicon-based techniques, machine learning techniques, deep learning techniques, and hybrid techniques. It discusses the importance of these techniques based on different aspects such as their impact on the issues of SA, levels of analysis, and performance evaluation measures. The research puts forward a comprehensive overview of the majority of the work done in Hindi SA. This study will help researchers in finding out resources such as annotated datasets, linguistic resources, and lexical resources. This survey delivers some significant findings and presents overall future research directions in the field of Hindi SA.

Management of menstrual hygiene, practices and perceptions among vulnerable Bharia women in Madhya Pradesh: A pilot survey

India has the highest proportion of adolescents and the highest adolescent pregnancy and childbirth rate in the tribal segment of the population. Very few studies have focused on the use of sanitary pad and quality of health care as menstrual hygiene practices. The data was collected among ‘Bharia’ women who identified as one of the particularly vulnerable tribal groups (PVTGs) in Madhya Pradesh. Towards comparing the use, source, and components of menstrual hygiene practices among adolescents and adult mothers, the sample included adolescent (10-19 years) and adult women 20-49 years of age. The relevant information was collected by trained female investigator with designed questionnaire from the respondents. Findings revealed that only 22.7% women are using sanitary pad during menstruation period in the studied tribe. It was seen the use of this absorbent (sanitary pad) for menstrual hygiene among the adolescents were found significantly higher (37.5%) than that of adult women (14.3%), whereas, the use of old cotton was found significantly higher (78.6%) among adult women. Awareness on menstrual hygiene about 50% adolescents and 64% adult women believed it is a curse of God. About 25% adolescents and 21.4% adult women believed menstruation is a natural process. The tribe has underprivileged menstrual hygiene practices owing to low awareness, illiteracy, and poverty. The special IEC education campaign with suitable strategy can be forced to optimum hygiene practices.

What is the current state of knowledge and distribution of evidence on established policies and programs to regulate international wildlife trade? A systematic map protocol

Background: The international trade of wildlife (animals and plants) provides critical resources for human communities worldwide and contributes to local, national, and international economies. However, increasing demand presents a significant threat to both species and ecosystems as well as wildlife-centered livelihoods. Concerns regarding illicit trade of wildlife and unsustainable harvest has propelled international wildlife trade regulation to the top of political and conservation agendas. Consequently, a broad range of interventions have been established to regulate the trade and address biodiversity decline. To gain a more comprehensive understanding of the impacts of international wildlife trade interventions, this protocol sets out the parameters for a systematic map which will comprehensively collate and describe the extent and distribution of the evidence base. The resulting map aims to provide insight to guide future research and inform practice. Methods: This systematic map will identify, map, and characterize the available evidence on the impacts of established policies and programs to address international wildlife trade. Specifically, the systematic map will describe: (1) the volume and distribution of studies that have examined impacts of various interventions on conservation, biological, and socioeconomic outcomes; (2) research methodologies that have been used to evaluate impacts; (3) distribution of studies on particular taxa and geographical areas; and (4) identify evidence gaps in need of more research. We will search two publication databases and several organizational and topical websites for relevant published articles and grey literature. In addition, a call for literature will be issued among relevant research networks. The titles, abstracts, and full texts of captured studies will be assessed against inclusion criteria. Double screening will be carried out on a subset of studies to ensure consistency. Relevant information from studies will be extracted using an a priori codebook. The resulting map will consist of descriptive statistics, a heat map in the form of a matrix, and a narrative synthesis describing characteristics of included studies.

Observed Changes in Agroclimate Metrics Relevant for Specialty Crop Production in California

Every decade, a suite of standardized climatological metrics known as climate normals are updated, providing averages of temperature and precipitation data over the previous 30-year period. Although some of these climate normals are directly applicable to agricultural producers, there are additional agroclimate metrics calculated from meteorological data that provide physiologically relevant information for on-farm management decisions. In this study, we identified a suite of energy-based agroclimate metrics and calculated changes over the two most recent normal periods (1981–2010 and 1991–2020), focusing on specialty crop production regions in California. Observed changes in agroclimate metrics were largely consistent with broader global warming trends. While most metrics showed small changes between the two periods, during the 1991–2020 period, the last spring freeze occurred ~5 days earlier as compared to the 1981-2010 period, contributing to a >6 day longer frost-free period in the Sacramento and Salinas Valleys; likewise an additional 6.4 tropical nights (Tn > 20 °C) occurred in the Coachella Valley during the 1991-2020 period. A complementary trend analysis of the agroclimate metrics over the 1981–2020 period showed significant increases in growing degree days across all agricultural regions, while significant increases in heat exposure were found for the Salinas and Imperial Valleys and over the Central Coast region. Moreover, summer reference evapotranspiration increased approximately 40 mm in California’s Central Valley during 1981–2020, with implications for agricultural water resources. Quantifying the shifts in these agroclimate metrics between the two most recent 30-year normal periods and the accompanying 40-year trends provides context for understanding and communicating around changing climatic baselines and underscores the need for adaptation to meet the challenge that climate change poses to agriculture both in the future and in the present.

The role of Omani academic libraries in supporting the knowledge economy

This study aims to Identify the role of Academic Omani Libraries in supporting knowledge economy, through Bryson's five indicators appeared in 2001: library infrastructure, information services; activities carried out by the libraries for creative ideas, and for supporting innovation. In addition to building collections that are capable to new requirements of Knowledge management. The study adopted the questionnaire as a tool of the descriptive method to collect and analyze data. The results indicate that Omani academic libraries have a good infrastructure in communications and information technology that supports research, and facilitates the use of electronic services. Moreover, Library catalogues and databases have been indicated by respondents as sufficient tools for exploring relevant information, especially in libraries, which are continuously organizing training programs in new developments of knowledge economy. On the other hand, the results show weaknesses of Omani academic libraries in supporting creative ideas, and in encouraging their employees to find out creative solutions for various problems facing them. The study concluded with many recommendations, the most important of which are: the need, for Omani academic libraries, to keep up with developments in the fields of knowledge management and economy, and to support creative ideas through collaboration with local and international professional associations. In addition, to encourage creative thinking with collaboration with organizations concerned with innovation in the country.

Carpooling in Connected and Autonomous Vehicles: Current Solutions and Future Directions

Owing to the advancements in communication and computation technologies, the dream of commercialized connected and autonomous cars is becoming a reality. However, among other challenges such as environmental pollution, cost, maintenance, security, and privacy, the ownership of vehicles (especially for Autonomous Vehicles (AV)) is the major obstacle in the realization of this technology at the commercial level. Furthermore, the business model of pay-as-you-go type services further attracts the consumer because there is no need for upfront investment. In this vein, the idea of car-sharing ( aka carpooling) is getting ground due to, at least in part, its simplicity, cost-effectiveness, and affordable choice of transportation. Carpooling systems are still in their infancy and face challenges such as scheduling, matching passengers interests, business model, security, privacy, and communication. To date, a plethora of research work has already been done covering different aspects of carpooling services (ranging from applications to communication and technologies); however, there is still a lack of a holistic, comprehensive survey that can be a one-stop-shop for the researchers in this area to, i) find all the relevant information, and ii) identify the future research directions. To fill these research challenges, this paper provides a comprehensive survey on carpooling in autonomous and connected vehicles and covers architecture, components, and solutions, including scheduling, matching, mobility, pricing models of carpooling. We also discuss the current challenges in carpooling and identify future research directions. This survey is aimed to spur further discussion among the research community for the effective realization of carpooling.