The (real) need for a human touch: testing a human–machine hybrid topic classification workflow on a New York Times corpus

The classification of the items of ever-increasing textual databases has become an important goal for a number of research groups active in the field of computational social science. Due to the increased amount of text data there is a growing number of use-cases where the initial effort of human classifiers was successfully augmented using supervised machine learning (SML). In this paper, we investigate such a hybrid workflow solution classifying the lead paragraphs of New York Times front-page articles from 1996 to 2006 according to policy topic categories (such as education or defense) of the Comparative Agendas Project (CAP). The SML classification is conducted in multiple rounds and, within each round, we run the SML algorithm on n samples and n times if the given algorithm is non-deterministic (e.g., SVM). If all the SML predictions point towards a single label for a document, then it is classified as such (this approach is also called a “voting ensemble"). In the second step, we explore several scenarios, ranging from using the SML ensemble without human validation to incorporating active learning. Using these scenarios, we can quantify the gains from the various workflow versions. We find that using human coding and validation combined with an ensemble SML hybrid approach can reduce the need for human coding while maintaining very high precision rates and offering a modest to a good level of recall. The modularity of this hybrid workflow allows for various setups to address the idiosyncratic resource bottlenecks that a large-scale text classification project might face.

Theoretical aspects of research of infrastructure are in context of the project of approach

This article the infrastructure essence was investigated and the need exploring this category as an important factor of the state socio-economic development was underline. The developed infrastructure complex builds objective conditions to solve scientific and technical issues, optimizing economic ties, integrating the state into the global economic space. The implementation of projects the construction and modernization of infrastructure facilities will provide a multiplier effect and will contribute to economic growth. The relevance of the problems described is a condition for deep research in this direction. The purpose of the article is studeing the theoretical foundations of infrastructure and develops proposals for improving the classification of infrastructure in the context of the project approach. The survey methodology is based on methods of analysis, synthesis and comparison. These methods were used to study of existing approaches to the disclosure to essence of the infrastructure. The scientific novelty of the research is to develop a hierarchy of infrastructure classification within the framework of the project approach by grouping projects that ensure the functioning of the infrastructure, according to such hierarchical criteria as: the public purpose of the infrastructure project, the ability of the infrastructure project to generate cash flows, the goals of the infrastructure project. Conclusions. Investigations have show that in the economic literature there were different approaches to understand the essence of infrastructure, determining elements and classification one. The reasons for the different opinions lie in the practical using of the definition of infrastructure in various industries, spheres and areas of scientific research. Following investigate which had found out that the infrastructure was a multi-level system, which consists of material and non-material objects. Therefore, it is necessary to develop a classification for systematizing them according to different criteria Key words: infrastructure, infrastructure classification, project approach, infrastructure project, types of infrastructure.

Locally projective graphs and their densely embedded subgraphs

The article contributes to the classification project of locally projective graphs and their locally projective groups of automorphisms outlined in Chapter 10 of Ivanov (The Mathieu Groups, Cambridge University Press, Cambridge, 2018). We prove that a simply connected locally projective graph $$\Gamma $$ Γ of type (n, 3) for $$n \ge 3$$ n ≥ 3 contains a densely embedded subtree provided (a) it contains a (simply connected) geometric subgraph at level 2 whose stabiliser acts on this subgraph as the universal completion of the Goldschmidt amalgam $$G_3^1\cong \{S_4 \times 2,S_4 \times 2\}$$ G 3 1 ≅ { S 4 × 2 , S 4 × 2 } having $$S_6$$ S 6 as another completion, (b) for a vertex x of $$\Gamma $$ Γ the group $$G_{\frac{1}{2}}(x)$$ G 1 2 ( x ) which stabilizes every line passing through x induces on the neighbourhood $$\Gamma (x)$$ Γ ( x ) of x the (dual) natural module $$2^n$$ 2 n of $$G(x)/G_{\frac{1}{2}}(x) \cong L_n(2)$$ G ( x ) / G 1 2 ( x ) ≅ L n ( 2 ) , (c) G(x) splits over $$G_{\frac{1}{2}}(x)$$ G 1 2 ( x ) , (d) the vertex-wise stabilizer $$G_1(x)$$ G 1 ( x ) of the neighbourhood of x is a non-trivial group, and (e) $$n \ne 4$$ n ≠ 4 .

A Blended Artificial Intelligence Approach for Spectral Classification of Stars in Massive Astronomical Surveys

This paper analyzes and compares the sensitivity and suitability of several artificial intelligence techniques applied to the Morgan–Keenan (MK) system for the classification of stars. The MK system is based on a sequence of spectral prototypes that allows classifying stars according to their effective temperature and luminosity through the study of their optical stellar spectra. Here, we include the method description and the results achieved by the different intelligent models developed thus far in our ongoing stellar classification project: fuzzy knowledge-based systems, backpropagation, radial basis function (RBF) and Kohonen artificial neural networks. Since one of today’s major challenges in this area of astrophysics is the exploitation of large terrestrial and space databases, we propose a final hybrid system that integrates the best intelligent techniques, automatically collects the most important spectral features, and determines the spectral type and luminosity level of the stars according to the MK standard system. This hybrid approach truly emulates the behavior of human experts in this area, resulting in higher success rates than any of the individual implemented techniques. In the final classification system, the most suitable methods are selected for each individual spectrum, which implies a remarkable contribution to the automatic classification process.

Vegetation classification enables inferring mesoscale spatial variation in plant invasibility

Large-scale control of invasive plants can benefit strongly from reliable assessment of spatial variation in plant invasibility. With this knowledge, limited management resources can be concentrated in areas of high invasion risk. We assessed the influence of spatial environments and proximity to roads on the invasibility of African mustard (Brassica tournefortii Gouan) over the 280,000-ha Barry M. Goldwater Range West in southwestern Arizona, USA. We used presence/absence data of B. tournefortii acquired from a vegetation classification project, in which lands were mapped to the level of vegetation subassociations. Logistic regression models suggested that spatial environments represented by the subassociations, not proximity to roads, represented the only factor significantly explaining B. tournefortii presence. We then used the best model to predict B. tournefortii invasibility in each subassociation. This prediction indicates management strategy should differ between the western part and the central to eastern part of the range. The western range is a large spatial continuum with intermediate to high invasion risk, vulnerable to an untethered spread of B. tournefortii. Controlling efforts should focus on preventing existing local populations from further expansion. The central and eastern ranges are a mosaic varying strongly in invasion risk. Control efforts can take advantage of natural invasion barriers and further reduce connectivity through removal of source populations connected with other high-risk locations via roads and other dispersal corridors. We suggest our approach as one effective way to combine vegetation classification and plant invasion assessment to manage complex landscapes over large ranges, especially when this approach is used through an iterative prediction–validation process to achieve adaptive management of invasive plants.