Open Access

Background Information about the global structure of agriculture and nutrient production and its diversity is essential to improve present understanding of national food production patterns, agricultural livelihoods, and food chains, and their linkages to land use and their associated ecosystems services. Here we provide a plausible breakdown of global agricultural and nutrient production by farm size, and also study the associations between farm size, agricultural diversity, and nutrient production. This analysis is crucial to design interventions that might be appropriately targeted to promote healthy diets and ecosystems in the face of population growth, urbanisation, and climate change. Methods We used existing spatially-explicit global datasets to estimate the production levels of 41 major crops, seven livestock, and 14 aquaculture and fish products. From overall production estimates, we estimated the production of vitamin A, vitamin B12, folate, iron, zinc, calcium, calories, and protein. We also estimated the relative contribution of farms of different sizes to the production of different agricultural commodities and associated nutrients, as well as how the diversity of food production based on the number of different products grown per geographic pixel and distribution of products within this pixel (Shannon diversity index [H]) changes with different farm sizes. Findings Globally, small and medium farms (≤50 ha) produce 51–77% of nearly all commodities and nutrients examined here. However, important regional differences exist. Large farms (>50 ha) dominate production in North America, South America, and Australia and New Zealand. In these regions, large farms contribute between 75% and 100% of all cereal, livestock, and fruit production, and the pattern is similar for other commodity groups. By contrast, small farms (≤20 ha) produce more than 75% of most food commodities in sub-Saharan Africa, southeast Asia, south Asia, and China. In Europe, west Asia and north Africa, and central America, medium-size farms (20–50 ha) also contribute substantially to the production of most food commodities. Very small farms (≤2 ha) are important and have local significance in sub-Saharan Africa, southeast Asia, and south Asia, where they contribute to about 30% of most food commodities. The majority of vegetables (81%), roots and tubers (72%), pulses (67%), fruits (66%), fish and livestock products (60%), and cereals (56%) are produced in diverse landscapes (H>1·5). Similarly, the majority of global micronutrients (53–81%) and protein (57%) are also produced in more diverse agricultural landscapes (H>1·5). By contrast, the majority of sugar (73%) and oil crops (57%) are produced in less diverse ones (H≤1·5), which also account for the majority of global calorie production (56%). The diversity of agricultural and nutrient production diminishes as farm size increases. However, areas of the world with higher agricultural diversity produce more nutrients, irrespective of farm size. Interpretation Our results show that farm size and diversity of agricultural production vary substantially across regions and are key structural determinants of food and nutrient production that need to be considered in plans to meet social, economic, and environmental targets. At the global level, both small and large farms have key roles in food and nutrition security. Efforts to maintain production diversity as farm sizes increase seem to be necessary to maintain the production of diverse nutrients and viable, multifunctional, sustainable landscapes. Funding Commonwealth Scientific and Industrial Research Organisation, Bill & Melinda Gates Foundation, CGIAR Research Programs on Climate Change, Agriculture and Food Security and on Agriculture for Nutrition and Health funded by the CGIAR Fund Council, Daniel and Nina Carasso Foundation, European Union, International Fund for Agricultural Development, Australian Research Council, National Science Foundation, Gordon and Betty Moore Foundation, and Joint Programming Initiative on Agriculture, Food Security and Climate Change—Belmont Forum.

This dissertation addresses the computational modeling of situation entity types (Smith, 2003), an inventory of clause types capturing aspectual and semantic distinctions that are relevant for various natural language processing tasks including temporal discourse processing and information extraction. The focus of our work is on automatically identifying the situation entity types STATE, ("John is tall"), EVENT ("John won the race"), GENERALIZING SENTENCE ("John cycles to work") and GENERIC SENTENCE ("Elephants are mammals"). We create a large corpus of texts from a variety of genres and domains, annotating each clause with its situation entity type and with linguistic phenomena that we identify as relevant for distinguishing the types. Specifically, we mark each clause with its lexical aspectual class, which takes the values stative ("be," "know") or dynamic ("run," "win"), and whether the clause is episodic or habitual, i.e., whether it refers to a particular event or whether it generalizes over situations. In addition, we annotate whether a clause's subject is generic or not, i.e., whether it refers to a kind ("dogs") or to a particular individual ("my dog"). Our human annotators achieve substantial agreement for all of these annotation tasks. Based on this corpus, we conduct a detailed corpus-linguistic study of situation entity type distributions and variation in inter-annotator agreement depending on the genre. In the second part of this dissertation, we create computational models for each of the above mentioned classification tasks in a supervised setting, advancing the state-of-the-art in each case. We find a range of syntactic-semantic features including distributional information and corpus-based linguistic indicators to be helpful. Using a sequence labeling method, we are able to leverage discourse information in order to improve the recognition of genericity, which often cannot be decided without taking the sentences in the context into account. We show our models to perform robustly across domains. Our publicly available data set and implementation form the basis for future research on situation entity types and related aspectual phenomena, among others as a preprocessing step into various natural language processing tasks.

This paper addresses the automatic recognition of telicity, an aspectual notion. A telic event includes a natural endpoint (“she walked home”), while an atelic event does not (“she walked around”). Recognizing this difference is a prerequisite for temporal natural language understanding. In English, this classification task is difficult, as telicity is a covert linguistic category. In contrast, in Slavic languages, aspect is part of a verb’s meaning and even available in machine-readable dictionaries. Our contributions are as follows. We successfully leverage additional silver standard training data in the form of projected annotations from parallel English-Czech data as well as context information, improving automatic telicity classification for English significantly compared to previous work. We also create a new data set of English texts manually annotated with telicity.

Der Textildiscounter KiK wird derzeit vor dem Landgericht Dortmund mit der Forderung konfrontiert, den verheerenden Brand in einer pakistanischen Textilfabrik, die für ihn fertigte, mitzuverantworten. Das Verfahren könnte paradigmatische Bedeutung für die Durchsetzung internationaler Unternehmensverantwortung in Deutschland erlangen. Der Beitrag nimmt dies zum Anlass, die einschlägigen völker-, unions- und bundesrechtlichen Rechtsgrundlagen zu systematisieren und Möglichkeiten zu ihrer Weiterentwicklung aufzuzeigen.

Das Gesellschaftsrecht gehört zu den wichtigen Nebengebieten in der Vorbereitung auf beide Staatsexamina. Aus Sicht des Klausurenerstellers lässt es sich mit zahlreichen Fragestellungen aus den ersten drei Büchern des BGB verknüpfen. Der folgende Fortsetzungsbeitrag widmet sich in einem Überblick den gängigen Examensproblemen des Gesellschaftsrechts.

This paper describes our contribution to the closed track of the Shared Task Translation Inference across Dictionaries (TIAD2017), 1 held in conjunction with the first Conference on Language Data and Knowledge (LDK-2017). In our approach, we use supervised machine learning to predict high-quality candidate translation pairs. We train a Support Vector Machine using several features, mostly of the translation graph, but also taking into consideration string similarity (Levenshtein distance). As the closed track does not provide manual training data, we define positive training examples as translation candidate pairs which occur in a cycle in which there is a direct connection.

The appearance of topological effects in systems exhibiting a nontrivial topological band structure strongly relies on the coherent wave nature of the equations of motion. Here, we reveal topological dynamics in a classical stochastic random walk version of the Su-Schrieffer-Heeger model with no relation to coherent wave dynamics. We explain that the commonly used topological invariant in the momentum space translates into an invariant in a counting-field space. This invariant gives rise to clear signatures of the topological phase in an associated escape time distribution.