The environmental impacts are commonly quantified in the EIA studies by rating, ranking and scaling. The National EIA Guidelines, 1993, Nepal provides a guideline to score the impacts in terms of magnitude, extent, an...The environmental impacts are commonly quantified in the EIA studies by rating, ranking and scaling. The National EIA Guidelines, 1993, Nepal provides a guideline to score the impacts in terms of magnitude, extent, and duration. This step is commonly known as impact prediction in the EIA process. The predicted scores are multiplied by the weightage value of the resource;likely to be affected. The application of the weightage transforms the predicted values of the impacts into their “significance”—a concept used in the environmental decision making. In other words the significance value entails assignment of relative judgment values to the impacts. The impacts, thus, can ranked based on their significance. The impact ranking is more useful in evaluating the socio-economic impacts. Unlike air, water and noise quality, which can be assessed against established standards;the socio-economic impacts do not have standard scale and are difficult to rank. Importance weighting of socio-economic impacts are commonly determined by the consensus obtained from the interaction with the local people, agencies, NGOs and experts. The impact ranking in the EIA process is unavoidable, firstly to prioritize the urgent environmental issues and design mitigation measures accordingly and also provide coherent linkages among the issues, and plan monitoring and auditing linkage with the proposed mitigation measures. Furthermore, it also provide strong basis of decision making, and thus facilitate the decision makers. The process of impact prediction, determination of significance and ranking were applied in the EIA of Indrwati-3 Hydroelectric Project, which is one of the successful cases of EIA in Nepal. The authors believe that the impacts predicted and quantified through this method are being focussed to more on the local concerns, since it seeks an active involvement of the local people who are likely to be affected.展开更多
领域自适应是解决低资源问题的一种通用方式,可应用于各种自然语言处理的任务中.当前针对命名实体识别(named entity recognition,NER)任务的领域自适应研究通常从单一的源领域迁移到目标领域,在目标领域和源领域相近的情况下,这种方式...领域自适应是解决低资源问题的一种通用方式,可应用于各种自然语言处理的任务中.当前针对命名实体识别(named entity recognition,NER)任务的领域自适应研究通常从单一的源领域迁移到目标领域,在目标领域和源领域相近的情况下,这种方式能够取得较好的识别效果,但是在目标领域与源领域相关度不高的情况下,单一领域迁移方式存在很大的局限性.针对这一问题,提出一种融合多源领域贡献度加权的自适应NER模型(multi-domain adaptation NER model based on importance weighting,MDAIW).1)通过多个领域的知识迁移来提升目标领域的实体识别性能;2)根据不同领域及其领域内样本对目标领域的重要性,计算领域贡献度;3)将领域贡献度引入到NER模型中,以此来实现更好的模型领域适应性.最终在多个目标领域上进行实验,性能皆优于当前性能最好的方法,验证了模型的有效性.展开更多
文摘The environmental impacts are commonly quantified in the EIA studies by rating, ranking and scaling. The National EIA Guidelines, 1993, Nepal provides a guideline to score the impacts in terms of magnitude, extent, and duration. This step is commonly known as impact prediction in the EIA process. The predicted scores are multiplied by the weightage value of the resource;likely to be affected. The application of the weightage transforms the predicted values of the impacts into their “significance”—a concept used in the environmental decision making. In other words the significance value entails assignment of relative judgment values to the impacts. The impacts, thus, can ranked based on their significance. The impact ranking is more useful in evaluating the socio-economic impacts. Unlike air, water and noise quality, which can be assessed against established standards;the socio-economic impacts do not have standard scale and are difficult to rank. Importance weighting of socio-economic impacts are commonly determined by the consensus obtained from the interaction with the local people, agencies, NGOs and experts. The impact ranking in the EIA process is unavoidable, firstly to prioritize the urgent environmental issues and design mitigation measures accordingly and also provide coherent linkages among the issues, and plan monitoring and auditing linkage with the proposed mitigation measures. Furthermore, it also provide strong basis of decision making, and thus facilitate the decision makers. The process of impact prediction, determination of significance and ranking were applied in the EIA of Indrwati-3 Hydroelectric Project, which is one of the successful cases of EIA in Nepal. The authors believe that the impacts predicted and quantified through this method are being focussed to more on the local concerns, since it seeks an active involvement of the local people who are likely to be affected.
文摘领域自适应是解决低资源问题的一种通用方式,可应用于各种自然语言处理的任务中.当前针对命名实体识别(named entity recognition,NER)任务的领域自适应研究通常从单一的源领域迁移到目标领域,在目标领域和源领域相近的情况下,这种方式能够取得较好的识别效果,但是在目标领域与源领域相关度不高的情况下,单一领域迁移方式存在很大的局限性.针对这一问题,提出一种融合多源领域贡献度加权的自适应NER模型(multi-domain adaptation NER model based on importance weighting,MDAIW).1)通过多个领域的知识迁移来提升目标领域的实体识别性能;2)根据不同领域及其领域内样本对目标领域的重要性,计算领域贡献度;3)将领域贡献度引入到NER模型中,以此来实现更好的模型领域适应性.最终在多个目标领域上进行实验,性能皆优于当前性能最好的方法,验证了模型的有效性.
