Logical table-to-text generation is a task that involves generating logically faithful sentences from tables, which requires models to derive logical level facts from table records via logical inference. It raises a new challenge on the logical-level content planning of table-to-text models. However, directly learning the logical inference knowledge from table-text pairs is very difficult for neural models because of the ambiguity of natural language and the scarcity of parallel data. Hence even large-scale pre-trained language models present low logical fidelity on logical table-to-text. In this work, we propose a PLOG (Pretrained Logical Form Generator) framework to improve the generation fidelity. Specifically, PLOG is first pretrained on a table-to-logic-form generation (table-to-logic) task, then finetuned on downstream table-to-text tasks. The formal definition of logical forms enables us to collect large amount of accurate logical forms from tables without human annotation. In addition, PLOG can learn logical inference from table-logic pairs much more definitely than from table-text pairs. To evaluate our model, we further collect a controlled logical table-to-text dataset CONTLOG based on an existing dataset. On two benchmarks, LOGICNLG and CONTLOG, PLOG outperforms strong baselines by a large margin on the logical fidelity, demonstrating the effectiveness of table-to-logic pretraining.

A risk assessment method is then used to identify the flood risk in each community using the following risk factors: the area's urbanized area ratio, literacy rate, mortality rate, poverty incidence, radio/TV penetration, and state of structural and nonstructural measures. Vulnerability is defined as a weighted-sum of these components. A „penalty‟ was imposed for reduced vulnerability. Optimization comparison was done with MatLab‟s Genetic Algorithms and Simulated Annealing; Results showed „extreme‟ solutions and realistic designs, for simulated annealing and genetic algorithm, respectively. INTRODUCTION Disaster Risk Management (DRM) at the local, regional, and global scale continues to generate great research interest of a complex, multidisciplinary nature, involving the interplay of scientific, social, economic, and political dimensions. Driven by the series of disasters of increasing frequency and magnitude, DRM meaning and context has evolved into an internationally accepted definition: a systemic approach to identifying, assessing and reducing risk of all kinds associated with hazards and human activities with identified operational and practical disaster risk reduction initiatives. These initiatives have been clarified by the international community through UN‟s 2005 World Conference on Disaster Reduction in Kobe, Japan and accepted as the DRR framework, known as the Hyogo Framework of Action [1]. The ultimate objective of all DRM initiatives remains simple: reduce the loss of lives and property, and improve the capacity of communities to cope with disasters. The 2005 Hyogo Framework of Action (HFA) has been used to review UN member states‟ respective DRM initiatives.