scholarly journals Data-to-text Generation with Macro Planning

2021 ◽  
Vol 9 ◽  
pp. 510-527
Author(s):  
Ratish Puduppully ◽  
Mirella Lapata

Abstract Recent approaches to data-to-text generation have adopted the very successful encoder-decoder architecture or variants thereof. These models generate text that is fluent (but often imprecise) and perform quite poorly at selecting appropriate content and ordering it coherently. To overcome some of these issues, we propose a neural model with a macro planning stage followed by a generation stage reminiscent of traditional methods which embrace separate modules for planning and surface realization. Macro plans represent high level organization of important content such as entities, events, and their interactions; they are learned from data and given as input to the generator. Extensive experiments on two data-to-text benchmarks (RotoWire and MLB) show that our approach outperforms competitive baselines in terms of automatic and human evaluation.

2020 ◽  
pp. 1-24
Author(s):  
Dequan Jin ◽  
Ziyan Qin ◽  
Murong Yang ◽  
Penghe Chen

We propose a novel neural model with lateral interaction for learning tasks. The model consists of two functional fields: an elementary field to extract features and a high-level field to store and recognize patterns. Each field is composed of some neurons with lateral interaction, and the neurons in different fields are connected by the rules of synaptic plasticity. The model is established on the current research of cognition and neuroscience, making it more transparent and biologically explainable. Our proposed model is applied to data classification and clustering. The corresponding algorithms share similar processes without requiring any parameter tuning and optimization processes. Numerical experiments validate that the proposed model is feasible in different learning tasks and superior to some state-of-the-art methods, especially in small sample learning, one-shot learning, and clustering.


Author(s):  
Oluwaseyi Feyisetan ◽  
Abhinav Aggarwal ◽  
Zekun Xu ◽  
Nathanael Teissier

Accurately learning from user data while ensuring quantifiable privacy guarantees provides an opportunity to build better ML models while maintaining user trust. Recent literature has demonstrated the applicability of a generalized form of Differential Privacy to provide guarantees over text queries. Such mechanisms add privacy preserving noise to vectorial representations of text in high dimension and return a text based projection of the noisy vectors. However, these mechanisms are sub-optimal in their trade-off between privacy and utility. In this proposal paper, we describe some challenges in balancing this trade-off. At a high level, we provide two proposals: (1) a framework called LAC which defers some of the noise to a privacy amplification step and (2), an additional suite of three different techniques for calibrating the noise based on the local region around a word. Our objective in this paper is not to evaluate a single solution but to further the conversation on these challenges and chart pathways for building better mechanisms.


10.2196/17687 ◽  
2020 ◽  
Vol 4 (8) ◽  
pp. e17687
Author(s):  
Kristina K Gagalova ◽  
M Angelica Leon Elizalde ◽  
Elodie Portales-Casamar ◽  
Matthias Görges

Background Integrated data repositories (IDRs), also referred to as clinical data warehouses, are platforms used for the integration of several data sources through specialized analytical tools that facilitate data processing and analysis. IDRs offer several opportunities for clinical data reuse, and the number of institutions implementing an IDR has grown steadily in the past decade. Objective The architectural choices of major IDRs are highly diverse and determining their differences can be overwhelming. This review aims to explore the underlying models and common features of IDRs, provide a high-level overview for those entering the field, and propose a set of guiding principles for small- to medium-sized health institutions embarking on IDR implementation. Methods We reviewed manuscripts published in peer-reviewed scientific literature between 2008 and 2020, and selected those that specifically describe IDR architectures. Of 255 shortlisted articles, we found 34 articles describing 29 different architectures. The different IDRs were analyzed for common features and classified according to their data processing and integration solution choices. Results Despite common trends in the selection of standard terminologies and data models, the IDRs examined showed heterogeneity in the underlying architecture design. We identified 4 common architecture models that use different approaches for data processing and integration. These different approaches were driven by a variety of features such as data sources, whether the IDR was for a single institution or a collaborative project, the intended primary data user, and purpose (research-only or including clinical or operational decision making). Conclusions IDR implementations are diverse and complex undertakings, which benefit from being preceded by an evaluation of requirements and definition of scope in the early planning stage. Factors such as data source diversity and intended users of the IDR influence data flow and synchronization, both of which are crucial factors in IDR architecture planning.


Author(s):  
Junchi Zhang ◽  
Yanxia Qin ◽  
Yue Zhang ◽  
Mengchi Liu ◽  
Donghong Ji

The task of event extraction contains subtasks including detections for entity mentions, event triggers and argument roles. Traditional methods solve them as a pipeline, which does not make use of task correlation for their mutual benefits. There have been recent efforts towards building a joint model for all tasks. However, due to technical challenges, there has not been work predicting the joint output structure as a single task. We build a first model to this end using a neural transition-based framework, incrementally predicting complex joint structures in a state-transition process. Results on standard benchmarks show the benefits of the joint model, which gives the best result in the literature.


Author(s):  
Yang Bai ◽  
Ziran Li ◽  
Ning Ding ◽  
Ying Shen ◽  
Hai-Tao Zheng

We study the problem of infobox-to-text generation that aims to generate a textual description from a key-value table. Representing the input infobox as a sequence, previous neural methods using end-to-end models without order-planning suffer from the problems of incoherence and inadaptability to disordered input. Recent planning-based models only implement static order-planning to guide the generation, which may cause error propagation between planning and generation. To address these issues, we propose a Tree-like PLanning based Attention Network (Tree-PLAN) which leverages both static order-planning and dynamic tuning to guide the generation. A novel tree-like tuning encoder is designed to dynamically tune the static order-plan for better planning by merging the most relevant attributes together layer by layer. Experiments conducted on two datasets show that our model outperforms previous methods on both automatic and human evaluation, and demonstrate that our model has better adaptability to disordered input.


2014 ◽  
Vol 3 (1) ◽  
pp. 50-56 ◽  
Author(s):  
Md. Shahidul Islam

Family planning programs are considered to be an important way to control the rapid population growth of Bangla-desh with the main focus being women. As a male dominant country, the knowledge of, attitude towards, and ap-proval of family planning is largely influenced by the male member of couples in their choice of appropriate contra-ceptive methods. This paper examined the determinants of current choices in family planning methods in relation to male knowledge, attitude and approval of family planning practices in Bangladesh. A total of 430 married men aged 15-49 years were interviewed in Narsingdi, a district town of Bangladesh. The findings revealed that the men’s level of contraceptive knowledge and their attitude to modern contraceptive was high in this area. The findings asserted that the contraceptive prevalence rate among couples was (62.1%), with oral pills (26.51%) and withdrawals (6.25%) being the most preferred modern and traditional methods respectively. The multinomial logistic regression model has been applied to understand the determinants of the choice of contraceptive method. These findings indi-cated that women in skilled occupation, positive attitude towards modern contraceptives of male, high level of knowledge on contraceptive methods of male, approval of family planning by male, and current living children had significantly more positive effects on using modern contraceptives by the couple. It was also found that couples who desired more children were less likely to use modern contraceptives. Alternatively, the education level of the hus-band and the desired number of additional children had a negative impact on the use of traditional methods while media exposure, a high level of knowledge on contraceptives, and an approval of family planning had positive im-pacts on the use of traditional contraceptives over not using any method. The government should increase the fund-ing and availability of family planning programs which promote the use of modern contraceptive methods, espe-cially those that are targeted towards the male population. South East Asia Journal of Public Health 2013; 3(1): 50-56 DOI: http://dx.doi.org/10.3329/seajph.v3i1.17711


Author(s):  
Максим Сергеевич Шевелин ◽  
Андрей Анатольевич Иванов ◽  
Александр Сергеевич Брежнев ◽  
Евгений Алексеевич Азаров

В статье представлены данные оригинального исследования по прогнозированию риска развития тромботических осложнений после хирургического лечения аневризм брюшного отдела аорты. С этой целью создана и реализована соответствующая программа исследования. В предоперационном периоде произведено обследование репрезентативной выборки тематических пациентов. Выделен комплекс факторов риска по развитию послеоперационных тромботических осложнений - тромбозов глубоких вен и тромбозов браншей протеза. В послеоперационном периоде определена фактическая вероятность их развития, выявлена диагностическая информативность (прогностическая значимость) оцениваемых периоперационных факторов риска. На основе наиболее значимых факторов разработана прогностическая математическая модель, позволяющая определять по их наличию и сочетанию высокий или низкий уровень риска развития осложнений. Работоспособность модели была проверена путем сравнения полученных клинических данных с результатами численного прогнозирования. Выявлен высокий уровень работоспособности. Верификация показала высокий уровень адекватности модели. Ее использование позволяет дифференцировать тематических пациентов на группы высокого, среднего и низкого риска по развитию послеоперационных тромботических осложнений. В соответствии с этим появилась возможность выбора рекомендации по использованию «стандартных» или «усиленных» программ антитромботической профилактики для каждого конкретного пациента. Полученные результаты имеют высокий уровень статистической значимости, что, в свою очередь, позволяет рекомендовать их к рассмотрению для использования в практике сосудистой хирургии на этапе планирования операций по поводу аневризм брюшного отдела аорты и составления программ профилактики послеоперационных тромботических осложнений The article presents data from an original study of predicting the risk of thrombotic complications after surgical treatment of abdominal aortic aneurysms. For this purpose, an appropriate research program has been created and implemented. In the preoperative period, a representative sample of thematic patients was examined. A complex of risk factors was identified for the development of postoperative thrombotic complications - deep vein thrombosis and thrombosis of prosthetics branches. In the postoperative period, the actual probability of their development was determined, diagnostic informativeness (prognostic value) of the estimated perioperative risk factors was revealed. Based on the most significant factors, a predictive mathematical model has been appearance that allows one to determine by their presence and combination a high, middle or low level of risk of complications. The performance of the model was tested by comparing the obtained clinical data with the results of numerical forecasting. A high level of performance has been identified. Verification showed a high level of model adequacy. Its use allows us to differentiate thematic patients into high and low risk groups according to the development of postoperative thrombotic complications. In accordance with this, it became possible to choose recommendations on the use of «standard» or «enhanced» antithrombotic prophylaxis programs for each specific patient. The results obtained have a high level of statistical significance, which, in turn, allows us to recommend them for consideration in the practice of vascular surgery at the planning stage of operations for abdominal aortic aneurysms and the development of programs for the prevention of postoperative thrombotic complications


Author(s):  
Eman A. Zabalawi ◽  
Abderazak Bakhouche ◽  
Randa El Chaar

The chapter covers practical risk managers' points during the planning stage of entrepreneurship to integrate into decision making and core business processes. The chapter includes three risks and how to improve the integration of risk management into organizational culture grouped into three high-level objectives. First, strategic risks include the study of competitors, macroeconomics with its industry changes. Second, financial risks such as the business return of investment and forecasting, customer payments, loan interest charges, and liquidity. Access to finance featured prominently in several studies as a constraint on SME development. The third is the operational risks and internal analysis that includes legal compliance, breakdown of essential equipment, employee mismatch, partnership, information technology, external events, and supply chain reliability. It is vital to establish an organizational culture where risk management is a daily component activity.


Molecules ◽  
2019 ◽  
Vol 24 (19) ◽  
pp. 3495
Author(s):  
Zhikai Zhang ◽  
Lihua Lin ◽  
Hongchi Tang ◽  
Shaowei Zeng ◽  
Yuan Guo ◽  
...  

A convenient and effective sucrose transport assay for Clostridium strains is needed. Traditional methods, such as 14C-sucrose isotope labelling, use radioactive materials and are not convenient for many laboratories. Here, a sucrose transporter from potato was introduced into Clostridium, and a fluorescence assay based on esculin was used for the analysis of sucrose transport in Clostridium strains. This showed that the heterologously expressed potato sucrose transporter is functional in Clostridium. Recombinant engineering of high-level sucrose transport would aid sucrose fermentation in Clostridium strains. The assay described herein provides an important technological platform for studying sucrose transporter function following heterologous expression in Clostridium.


2015 ◽  
Vol 1 (1) ◽  
pp. 49-51 ◽  
Author(s):  
Ajaya Kumar Dhakal ◽  
Sanjaya Dhakal

Medicine of present world demands high level of competency in both clinical examination and performing a procedure in patients. The traditional methods of bedside skill learning and teaching should be supplemented by instruction in clinical skills lab of basic important clinical skills. Every medical school should work towards establishment and incorporation of clinical skills lab in basics science subjects and clinical posting along with other subjects to make it Practice oriented and Student centred learning.Journal of Patan Academy of Health Sciences. 2014 Jun;1(1):49-51


Sign in / Sign up

Export Citation Format

Share Document