Computational Journalism: from Answering Question to Questioning Answers and Raising Good Questions

Our media is saturated with claims of ``facts'' made from data. Database research has in the past focused on how to answer queries, but has not devoted much attention to discerning more subtle qualities of the resulting claims, e.g., is a claim ``cherry-picking''? This paper proposes a Query Response Surface (QRS) based framework that models claims based on structured data as parameterized queries. A key insight is that we can learn a lot about a claim by perturbing its parameters and seeing how its conclusion changes. This framework lets us formulate and tackle practical fact-checking tasks --- reverse-engineering vague claims, and countering questionable claims --- as computational problems. Within the QRS based framework, we take one step further, and propose a problem along with efficient algorithms for finding high-quality claims of a given form from data, i.e. raising good questions, in the first place. This is achieved to using a limited number of high-valued claims to represent high-valued regions of the QRS. Besides the general purpose high-quality claim finding problem, lead-finding can be tailored towards specific claim quality measures, also defined within the QRS framework. An example of uniqueness-based lead-finding is presented for ``one-of-the-few'' claims, landing in interpretable high-quality claims, and an adjustable mechanism for ranking objects, e.g. NBA players, based on what claims can be made for them. Finally, we study the use of visualization as a powerful way of conveying results of a large number of claims. An efficient two stage sampling algorithm is proposed for generating input of 2d scatter plot with heatmap, evalutaing a limited amount of data, while preserving the two essential visual features, namely outliers and clusters. For all the problems, we present real-world examples and experiments that demonstrate the power of our model, efficiency of our algorithms, and usefulness of their results.

Ca2+ current and Ca(2+)-activated chloride current in isolated smooth muscle cells of the sheep urethra.

1. Isolated sheep urethral cells were studied using the perforated patch clamp technique (T = 37 degrees C). Depolarizing steps ranging from -40 to -10 mV evoked an inward current that peaked within 10 ms and a slower inward current. Stepping back to the holding potential of -80 mV evoked large inward tail currents. All three currents were abolished by nifedipine (1 microM). Substitution of external Ca2+ with Ba2+ resulted in potentiation of the fast inward current and blockade of the slow current and tails. 2. Changing the chloride equilibrium potential (ECl) from 0 to +27 mV shifted the reversal potential of the tail currents from 1 +/- 1 to 27 +/- 1 mV (number of cells, n = 5). Chloride channel blockers, niflumic acid (10 microM) and anthracene-9-carboxylic acid (9AC, 1 mM), reduced the slow current and tails suggesting that these were Ca(2+)-activated Cl- currents, ICl(Ca). 4. Caffeine (10 mM) induced currents that reversed at ECl and were blocked by niflumic acid (10 microM). 5. In current clamp mode, some cells developed spontaneous transient depolarizations (STDs) and action potentials. Short exposure to nifedipine blocked the action potentials and unmasked STDs. In contrast, 9AC and niflumic acid reduced the amplitude of the STDs and blocked the action potentials. 6. In conclusion, these cells have both L-type ICa and ICl(Ca). The former appears to be responsible for the upstroke of the action potential, while the latter may act as a pacemaker current.

'Trusting the numbers': mineral prospecting, raising finance and the governance of knowledge

Mineral prospecting and raising finance for ‘junior’ mining firms has historically been regarded as a speculative activity. For the regulators of securities markets upon which ‘junior’ mining companies seek to raise capital, a perennial problem has been handling not only the indeterminacy of scientific claims, but also the social basis of epistemic practices. This paper examines the production of a system of public warrant and associated knowledge practices intended to enable investors to differentiate between ‘destructive’ and ‘productive’ varieties of financial speculation. It traces the use of the notion of ‘disclosure’ in constructing and legitimizing the ‘juniors’ market in Canada. It argues that though the work of ‘economics’ may be necessary in the construction of markets, it is by no means sufficient. Attention must also be given to the ways in which legal models of ‘the free-market’ can be translated and constantly re-worked across the sites and spaces of regulatory practice, animating the geographies of markets.

Comparison of two assays, a faecal egg count reduction test (FECRT) and a coproantigen reduction test (CRT), for the diagnosis of resistance to triclabendazole in Fasciola hepatica in sheep

A sheep trial was performed to evaluate two diagnostic assays, a faecal egg count reduction test (FECRT) and a coproantigen reduction test (CRT), for the diagnosis of resistance of Fasciola hepatica to triclabendazole (TCBZ). The FECRT defines successful TCBZ treatment as a 95% or greater reduction in fluke faecal egg counts (FECs) at 14 days post-treatment (dpt). The CRT defines effective TCBZ treatment as faeces negative for Fasciola coproantigens at 14 dpt, as measured by the commercial BIO K201 coproantigen ELISA (Bio-X Diagnostics, Jemelle, Belgium).

Anthelmintic resistance in Northern Ireland (II): Variations in nematode control practices between lowland and upland sheep flocks

A questionnaire to obtain information on nematode control practices and sheep management was sent to over 1000 farmers in Northern Ireland. Replies were received from 305 flock owners, and data from 252 of them were analysed. Farms were divided into lowland and upland areas. Sizes of pasture and stocking rates on lowland and upland farms were 59.5 hectares, 6.99 sheep/hectare and 62.9 hectares and 10.01 sheep/hectare, respectively. Mean drenching rates for lambs and adults were 2.33 and 2.44, respectively, in lowland flocks and 2.73 and 2.71, respectively, in upland flocks. Between 2008 and 2011, the most frequently identified compounds in use were benzimidazoles and moxidectin in lowland flocks, and benzimidazoles and avermectins in upland flocks. Over the same period the most frequently identified commercial formulations were Tramazole (R), Panacur (R) and Allverm (R) (white drench), Levacide (R) (yellow drench), Oramec (R) (clear drench; avermectin), Cydectin (R) (clear drench; moxidectin) and Monepantel (R) (orange drench).

Most respondents (56.35%) treated their lambs at weaning and the most common time to treat ewes was identified to be pre-mating (67.86% of respondents).

The results of the questionnaire survey revealed that lowland annual drench frequency was 233 and 2.44 in lambs and ewes, respectively, although drench frequencies were higher in upland flocks: 2.73 and 2.71 for lambs and ewes, respectively.

Annual drench rotation was practiced by 43.96% of flock owners, but whether this was true rotation or pseudo-rotation (i.e., substitution of one anthelmintic product by another product belonging to the same chemical group of anthelmintics) could not be explicitly determined. (C) 2012 Elsevier B.V. All rights reserved.

'Trusting the numbers': mineral prospecting, raising finance and the governance of knowledge

Mineral prospecting and raising finance for ‘junior’ mining firms has historically been regarded as a speculative activity. For the regulators of securities markets upon which ‘junior’ mining companies seek to raise capital, a perennial problem has been handling not only the indeterminacy of scientific claims, but also the social basis of epistemic practices. This paper examines the production of a system of public warrant and associated knowledge practices intended to enable investors to differentiate between ‘destructive’ and ‘productive’ varieties of financial speculation. It traces the use of the notion of ‘disclosure’ in constructing and legitimizing the ‘juniors’ market in Canada. It argues that though the work of ‘economics’ may be necessary in the construction of markets, it is by no means sufficient. Attention must also be given to the ways in which legal models of ‘the free-market’ can be translated and constantly re-worked across the sites and spaces of regulatory practice, animating the geographies of markets.

Fingerprint Volatile Organic Compounds (VOC) analysis in Goat’s Milk and Halloumi Cheese as Marker for Authentication of Region of Origin

Goats’ milk is responsible for unique traditional products such as Halloumi cheese. The characteristics of Halloumi depend on the original features of the milk and on the conditions under which the milk has been produced such as feeding regime of the animals or region of production. Using a range of milk (33) and Halloumi (33) samples collected over a year from three different locations in Cyprus (A, Anogyra; K, Kofinou; P, Paphos), the potential for fingerprint VOC analysis as marker to authenticate Halloumi was investigated. This unique set up consists of an in-injector thermo desorption (VOCtrap needle) and a chromatofocusing system based on mass spectrometry (VOCscanner). The mass spectra of all the analyzed samples are treated by multivariate analysis (Principle component analysis and Discriminant functions analysis). Results showed that the highland area of product (P) is clearly identified in milks produced (discriminant score 67%). It is interesting to note that the higher similitude found on milks from regions “A” and “K” (with P being distractive; discriminant score 80%) are not ‘carried over’ on the cheeses (higher similitude between regions “A” and “P”, with “K” distinctive). Data have been broken down into three seasons. Similarly, the seasonality differences observed in different milks are not necessarily reported on the produced cheeses. This is expected due to the different VOC signatures developed in cheeses as part of the numerous biochemical changes during its elaboration compared to milk. VOC however it is an additional analytical tool that can aid in the identification of region origin in dairy products.