scholarly journals A Rapid Review of Image Captioning

2021 ◽  
Vol 6 (2) ◽  
pp. 158-169
Author(s):  
Adriyendi Adriyendi
Keyword(s):  
Process Model ◽  
Language Model ◽  
Human Vision ◽  
Rapid Review ◽  
Future Research ◽  
Automatic Process ◽  
Image Captioning ◽  
Input Model ◽  
Cross Language ◽  
Image Caption

Image captioning is an automatic process for generating text based on the content observed in an image. We do review, create framework, and build application model. We review image captioning into 4 categories based on input model, process model, output model, and lingual image caption. Input model is based on criteria caption, method, and dataset. Process model is based on type of learning, encoder-decoder, image extractor, and metric evaluation. Output model based on architecture, features extraction, feature aping, model, and number of caption. Lingual image caption based on language model with 2 groups: bilingual image caption and cross-language image caption. We also design framework with 3 framework model. Furthermore, we also build application with 3 application models. We also provide research opinions on trends and future research that can be developed with image caption generation. Image captioning can be further developed on computer vision versus human vision.

scholarly journals Image Caption Description of Traffic Scene Based on Deep Learning

2018 ◽  
Vol 36 (3) ◽  
pp. 522-527
Author(s):  
Shiru Qu ◽  
Yuling Xi ◽  
Songtao Ding
Keyword(s):  
Neural Network ◽  
Language Model ◽  
Attention Mechanism ◽  
Light Change ◽  
Generation Model ◽  
Semantic Description ◽  
Image Captioning ◽  
Benchmark Datasets ◽  
Abnormal Weather ◽  
Image Caption

It is a hard issue to describe the complex traffic scene accurately in computer vision. The traffic scene is changeable, which causes image captioning easily interfered by light changes and object occlusion. To solve this problem, we propose an image caption generation model based on attention mechanism. Combining convolutional neural network (CNN) and recurrent neural network (RNN) to generate an end-to-end description for traffic images. To generate a semantic description with distinct degree of discrimination, the attention mechanism is applied to language model. Using Flickr8K、Flickr30K and MS COCO benchmark datasets to validate the effectiveness of our method. The accuracy is promoted maximally by 8.6%, 12.4%, 19.3% and 21.5% in different evaluation metrics. Experiments show that our algorithm has good robustness in four different complex traffic scenarios, such as light change, abnormal weather environment, road marked target and various kinds of transportation tools.

scholarly journals Using a Process Approach to Pandemic Planning: A Case Study

2021 ◽  
Vol 11 (9) ◽  
pp. 4121
Author(s):  
Hana Tomaskova ◽  
Erfan Babaee Tirkolaee
Keyword(s):  
Process Model ◽  
Process Mining ◽  
Process Approach ◽  
Research Topic ◽  
Future Research ◽  
Critical Situation ◽  
Pandemic Planning ◽  
The Difference ◽  
Smart Technologies

The purpose of this article was to demonstrate the difference between a pandemic plan’s textual prescription and its effective processing using graphical notation. Before creating a case study of the Business Process Model and Notation (BPMN) of the Czech Republic’s pandemic plan, we conducted a systematic review of the process approach in pandemic planning and a document analysis of relevant public documents. The authors emphasized the opacity of hundreds of pages of text records in an explanatory case study and demonstrated the effectiveness of the process approach in reengineering and improving the response to such a critical situation. A potential extension to the automation and involvement of SMART technologies or process optimization through process mining techniques is presented as a future research topic.

scholarly journals Community intervention strategies to reduce the impact of financial strain and promote financial well-being: a comprehensive rapid review

2021 ◽  
Vol 28 (1) ◽  
pp. 42-50
Author(s):  
Nicole M. Glenn ◽  
Lisa Allen Scott ◽  
Teree Hokanson ◽  
Karla Gustafson ◽  
Melissa A. Stoops ◽  
...  
Keyword(s):  
Quality Of Life ◽  
Community Intervention ◽  
Well Being ◽  
Small Sample ◽  
Financial Strain ◽  
Rapid Review ◽  
Research Centre ◽  
Future Research ◽  
The Impact

Financial well-being describes when people feel able to meet their financial obligations, feel financially secure and are able to make choices that benefit their quality of life. Financial strain occurs when people are unable to pay their bills, feel stressed about money and experience negative impacts on their quality of life and health. In the face of the global economic repercussions of the COVID-19 pandemic, community-led approaches are required to address the setting-specific needs of residents and reduce the adverse impacts of widespread financial strain. To encourage evidence-informed best practices, a provincial health authority and community-engaged research centre collaborated to conduct a rapid review. We augmented the rapid review with an environmental scan and interviews. Our data focused on Western Canada and was collected prior to the pandemic (May–September 2019). We identified eight categories of community-led strategies to promote financial well-being: systems navigation and access; financial literacy and skills; emergency financial assistance; asset building; events and attractions; employment and educational support; transportation; and housing. We noted significant gaps in the evidence, including methodological limitations of the included studies (e.g. generalisability, small sample size), a lack of reporting on the mechanisms leading to the outcomes and evaluation of long-term impacts, sparse practice-based data on evaluation methods and outcomes, and limited intervention details in the published literature. Critically, few of the included interventions specifically targeted financial strain and/or well-being. We discuss the implications of these gaps in addition to possibilities and priorities for future research and practice. We also consider the results in relation to the COVID-19 pandemic and its economic consequences.

Prosody, productivity and word structure: the stød pattern of Modern Danish

2003 ◽  
Vol 26 (1) ◽  
pp. 5-44 ◽  
Author(s):  
Hans Basbøll
Keyword(s):  
Language Model ◽  
New Model ◽  
Word Structure ◽  
Cross Language ◽  
Complex Distribution

Parts of a new model of phonology-morphology-lexicon interplay is presented to account for the complex distribution of the Modern Danish stød (a syllabic prosody). Stød, which is sometimes productive for speakers, is analysed as a signal of the second mora of bimoraic syllables not subject to the Non-Stød Principle (NSP). The author's cross-language model for Systematically Graded Productivity of Endings (section 3) is shown to account for the application of NSP (section 4), and a detailed typology of lexemes with respect to stød-alternations, derived from the model, is presented in section 5. In section 6, a simple case of stød-alternations in inflection, viz. regular plurals of nouns, is given, and section 7 exemplifies stød and non-stød as a key to morphology for the addressee.

scholarly journals Pain management programmes via video conferencing: a rapid review

2020 ◽  
Vol 0 (0) ◽  
Author(s):  
Jackie Walumbe ◽  
Joletta Belton ◽  
Diarmuid Denneny
Keyword(s):  
Pain Management ◽  
Methodological Quality ◽  
Peer Group ◽  
Video Conferencing ◽  
Mesh Term ◽  
Rapid Review ◽  
Future Research ◽  
Chronic Pain Conditions ◽  
Pain Services

AbstractObjectivesDuring the current COVID-19 pandemic, healthcare has been transformed by the rapid switch from in person care to use of remote consulting, including video conferencing technology. Whilst much has been published on one-to-one video consultations, little literature exists on use of this technology to facilitate group interventions. Group pain management programmes are a core treatment provided by many pain services. This rapid review aimed to identify the extent of use of video conferencing technology for delivery of group pain management programmes and provide an overview of its use.MethodsA rapid review of the literature published up to April 2020 (PubMed, PsycINFO and PEDro) was performed. The search string consisted of three domains: pain/CP (MeSH term) AND Peer group[MeSH] AND Videoconferencing[MeSH]/Telemedicine[MeSH]/Remote Consultation[MeSH]. The studies were of poor methodological quality and study design, and interventions and chronic pain conditions were varied.ResultsLiterature searching yielded three eligible papers for this review. All studies had low methodological quality and risk of bias. Heterogeneity and variability in outcome reporting did not allow any pooling of data. The results demonstrated that videoconferencing for delivery of group programmes is possible, yet there is little extant literature on how to develop, deliver and measure outcomes of such programmes.ConclusionsThis review demonstrates that there is little evidence to support or guide the use of synchronous videoconferencing to deliver pain management programmes. We present issues to consider, informed by this review and our experience, when implementing video conferencing. Study quality of existing work is variable, and extensive future research is necessary.

Holistic Leader(ship) Development: An Integrative Process Model of Leader and Leadership Development

2021 ◽  
Author(s):  
H. Michael Schwartz ◽  
Pooja Khatija ◽  
Diana Bilimoria
Keyword(s):  
Leadership Development ◽  
Identity Development ◽  
Organizational Support ◽  
Process Model ◽  
Leader Development ◽  
Leadership Capacity ◽  
Future Research ◽  
Organizational Contexts ◽  
Leader Identity ◽  
Capital Development

The question of how to efficiently, holistically, and successfully develop leaders has been the focus of scholars and practitioners for several decades. Embedding the process of leader development in organizational contexts allows participants to develop and apply leadership knowledge, skills, and identity awareness. Embeddedness facilitates the holistic integration of the interactive processes of leader development (which focuses on increasing the leadership capacity of an individual) and leadership development (which focuses on increasing the leadership capacity of an organization), which is referred to in this article as leader(ship) development (LD). Two sub-processes involved in LD (i.e., general and situational identity development and knowledge/skill/social capital development) and four mechanisms of embeddedness that facilitate holistic LD (i.e., leader identity integration, opportunities to learn and develop in the organization, organizational support and feedback, and helping relationships) will be described. A discussion on the ways by which management education pedagogy can integrate and facilitate embeddedness and provide guidance for future research will follow.

Improved Framework using Rider Optimization Algorithm for Precise Image Caption Generation

2021 ◽  
pp. 2250021
Author(s):  
Chaitrali Prasanna Chaudhari ◽  
Satish Devane
Keyword(s):  
Language Processing ◽  
Optimization Algorithm ◽  
Short Term Memory ◽  
Activation Function ◽  
Input Image ◽  
Image Captioning ◽  
Word Generation ◽  
Adopted Model ◽  
Conventional Models ◽  
Image Caption

“Image Captioning is the process of generating a textual description of an image”. It deploys both computer vision and natural language processing for caption generation. However, the majority of the image captioning systems offer unclear depictions regarding the objects like “man”, “woman”, “group of people”, “building”, etc. Hence, this paper intends to develop an intelligent-based image captioning model. The adopted model comprises of few steps like word generation, sentence formation, and caption generation. Initially, the input image is subjected to the Deep learning classifier called Convolutional Neural Network (CNN). Since the classifier is already trained in the relevant words that are related to all images, it can easily classify the associated words of the given image. Further, a set of sentences is formed with the generated words using Long-Short Term Memory (LSTM) model. The likelihood of the formed sentences is computed using the Maximum Likelihood (ML) function, and the sentences with higher probability are taken, which is further used for generating the visual representation of the scene in terms of image caption. As a major novelty, this paper aims to enhance the performance of CNN by optimally tuning its weight and activation function. This paper introduces a new enhanced optimization algorithm Rider with Randomized Bypass and Over-taker update (RR-BOU) for this optimal selection. In the proposed RR-BOU is the enhanced version of the Rider Optimization Algorithm (ROA). Finally, the performance of the proposed captioning model is compared over other conventional models with respect to statistical analysis.

Business process model patterns: state-of-the-art, research classification and taxonomy

2019 ◽  
Vol 25 (5) ◽  
pp. 972-994 ◽  
Author(s):  
Michael Fellmann ◽  
Agnes Koschmider ◽  
Ralf Laue ◽  
Andreas Schoknecht ◽  
Arthur Vetter
Keyword(s):  
Business Process ◽  
Process Model ◽  
Pattern Language ◽  
Future Research ◽  
Business Process Model ◽  
Content Type ◽  
Research Activities ◽  
Starting Point ◽  
Research Classification

Purpose Patterns have proven to be useful for documenting general reusable solutions to a commonly occurring problem. In recent years, several different business process management (BPM)-related patterns have been published. Despite the large number of publications on this subject, there is no work that provides a comprehensive overview and categorization of the published business process model patterns. The purpose of this paper is to close this gap by providing a taxonomy of patterns as well as a classification of 89 research works. Design/methodology/approach The authors analyzed 280 research articles following a structured iterative procedure inspired by the method for taxonomy development from Nickerson et al. (2013). Using deductive and inductive reasoning processes embedded in concurrent as well as joint research activities, the authors created a taxonomy of patterns as well as a classification of 89 research works. Findings In general, the findings extend the current understanding of BPM patterns. The authors identify pattern categories that are highly populated with research works as well as categories that have received far less attention such as risk and security, the ecological perspective and process architecture. Further, the analysis shows that there is not yet an overarching pattern language for business process model patterns. The insights can be used as starting point for developing such a pattern language. Originality/value Up to now, no comprehensive pattern taxonomy and research classification exists. The taxonomy and classification are useful for searching pattern works which is also supported by an accompanying website complementing the work. In regard to future research and publications on patterns, the authors derive recommendations regarding the content and structure of pattern publications.

scholarly journals Serendipity and its study

2014 ◽  
Vol 70 (6) ◽  
pp. 1015-1038 ◽  
Author(s):  
Allen Edward Foster ◽  
David Ellis
Keyword(s):  
Information Seeking ◽  
Process Model ◽  
Empirical Studies ◽  
Science Research ◽  
Empirical Literature ◽  
Future Research ◽  
Content Type ◽  
Information Studies ◽  
Information Encountering ◽  
Citation Classics

Purpose – The purpose of this paper is to explore the concept of serendipity and approaches to its study particularly in relation to information studies. Design/methodology/approach – The origins of the term serendipity are described and its elaboration as an exploratory and explanatory concept in science and the social sciences are outlined. The distinction between serendipity and serendipity pattern is explained and theoretical and empirical studies of both serendipity and the serendipity patterns are explored. The relationship between information encountering is described. Empirical studies of serendipity using Citation Classics and other research approaches in information studies are described. Findings – The discrepancy between occurrences of serendipity in studies using Citation Classics and reported serendipity in philosophy of science, research anecdotes, information encountering and information seeking by inter-disciplinary researchers is highlighted. A comparison between a process model of serendipity and serendipity as an emergent behavioural characteristic are indicates directions for future research. Originality/value – The paper provides and original synthesis of the theoretical and empirical literature on serendipity with particular reference to work in information studies and an indication of the methodological difficulties involved in its study.

scholarly journals BioALBERT: A Simple and Effective Pre-trained Language Model for Biomedical Named Entity Recognition

2020 ◽  
Author(s):  
Usman Naseem ◽  
Matloob Khushi ◽  
Vinay Reddy ◽  
Sakthivel Rajendran ◽  
Imran Razzak ◽  
...  
Keyword(s):  
State Of The Art ◽  
Language Model ◽  
Named Entity Recognition ◽  
Training Data ◽  
Entity Recognition ◽  
Future Research ◽  
Named Entity ◽  
Domain Specific ◽  
Context Dependent ◽  
Biomedical Named Entity Recognition

Abstract Background: In recent years, with the growing amount of biomedical documents, coupled with advancement in natural language processing algorithms, the research on biomedical named entity recognition (BioNER) has increased exponentially. However, BioNER research is challenging as NER in the biomedical domain are: (i) often restricted due to limited amount of training data, (ii) an entity can refer to multiple types and concepts depending on its context and, (iii) heavy reliance on acronyms that are sub-domain specific. Existing BioNER approaches often neglect these issues and directly adopt the state-of-the-art (SOTA) models trained in general corpora which often yields unsatisfactory results. Results: We propose biomedical ALBERT (A Lite Bidirectional Encoder Representations from Transformers for Biomedical Text Mining) - bioALBERT - an effective domain-specific pre-trained language model trained on huge biomedical corpus designed to capture biomedical context-dependent NER. We adopted self-supervised loss function used in ALBERT that targets on modelling inter-sentence coherence to better learn context-dependent representations and incorporated parameter reduction strategies to minimise memory usage and enhance the training time in BioNER. In our experiments, BioALBERT outperformed comparative SOTA BioNER models on eight biomedical NER benchmark datasets with four different entity types. The performance is increased for; (i) disease type corpora by 7.47% (NCBI-disease) and 10.63% (BC5CDR-disease); (ii) drug-chem type corpora by 4.61% (BC5CDR-Chem) and 3.89 (BC4CHEMD); (iii) gene-protein type corpora by 12.25% (BC2GM) and 6.42% (JNLPBA); and (iv) Species type corpora by 6.19% (LINNAEUS) and 23.71% (Species-800) is observed which leads to a state-of-the-art results. Conclusions: The performance of proposed model on four different biomedical entity types shows that our model is robust and generalizable in recognizing biomedical entities in text. We trained four different variants of BioALBERT models which are available for the research community to be used in future research.

Sign in / Sign up

Export Citation Format

Share Document