The complexity of narrative interferes in the use of conjunctions in children with specific language impairment

PURPOSE: To verify the use of conjunctions in narratives, and to investigate the influence of stimuli's complexity over the type of conjunctions used by children with specific language impairment (SLI) and children with typical language development. METHODS: Participants were 40 children (20 with typical language development and 20 with SLI) with ages between 7 and 10 years, paired by age range. Fifteen stories with increasing of complexity were used to obtain the narratives; stories were classified into mechanical, behavioral and intentional, and each of them was represented by four scenes. Narratives were analyzed according to occurrence and classification of conjunctions. RESULTS: Both groups used more coordinative than subordinate conjunctions, with significant decrease in the use of conjunctions in the discourse of SLI children. The use of conjunctions varied according to the type of narrative: for coordinative conjunctions, both groups differed only between intentional and behavioral narratives, with higher occurrence in behavioral ones; for subordinate conjunctions, typically developing children's performance did not show differences between narratives, while SLI children presented fewer occurrences in intentional narratives, which was different from other narratives. CONCLUSION: Both groups used more coordinative than subordinate conjunctions; however, typically developing children presented more conjunctions than SLI children. The production of children with SLI was influenced by stimulus, since more complex narratives has less use of subordinate conjunctions.

Complexity of inferences in polytree-shaped semi-qualitative probabilistic networks

Semi-qualitative probabilistic networks (SQPNs) merge two important graphical model formalisms: Bayesian networks and qualitative probabilistic networks. They provade a very Complexity of inferences in polytree-shaped semi-qualitative probabilistic networks and qualitative probabilistic networks. They provide a very general modeling framework by allowing the combination of numeric and qualitative assessments over a discrete domain, and can be compactly encoded by exploiting the same factorization of joint probability distributions that are behind the bayesian networks. This paper explores the computational complexity of semi-qualitative probabilistic networks, and takes the polytree-shaped networks as its main target. We show that the inference problem is coNP-Complete for binary polytrees with multiple observed nodes. We also show that interferences can be performed in time linear in the number of nodes if there is a single observed node. Because our proof is construtive, we obtain an efficient linear time algorithm for SQPNs under such assumptions. To the best of our knowledge, this is the first exact polynominal-time algorithm for SQPn. Together these results provide a clear picture of the inferential complexity in polytree-shaped SQPNs.

Catch rates of artisanal fisheries targeting Dasyatis guttata (Bloch & Schneider, 1801) in Todos os Santos Bay, northeastern Brazil

Dasyatis guttata has been target of artisanal fisheries in the coast of Bahia (Northeast Brazil) mainly by “arraieira” (gillnet) and “grozeira” (bottom long-line), but until now there is no stock assessment study. One of the important data for this knowledge is reliable indices of abundance. The aims of the present work are to: (1) estimate the best predictor for relative abundance (catch-per-unit-of-effort, CPUE), examining whether catch (production – kg) was related to: soak time of the gear, size of the gillnet or number of hooks, applying generalized linear model (GLM); (2) estimate the annual CPUE (kg/hooks and kg/m) averaged by gear; and (3) assess the temporal CPUE variance. Based on monthly sampling between January 2012 and January 2013, 222 landings by grozeira and 76 by arraiaiera were recorded in the two landing sites in Todos os Santos Bay, Bahia. A total of 14,550 kg (average = 44 kg/month) of D. guttata was captured. Models for both gears were highly significant (P < 0.0001). The analysis indicated that the most appropriate variable for CPUE analysis was the size of the gillnet (P < 0.001) and the number of hooks (P < 0.0001). Soak time of the gear was not significant for both gears (P = 0.4). High residual deviance expresses the complexity of the relations between ecosystem factors and other fisheries factors affecting relative abundance, which were not considered in this study. The average CPUE by grozeira was 6.39 kg/100 hooks ± 8.89 and by arraieira, 1.47 kg/100 m ± 1.66 over the year. Kruskal-Wallis test showed effect of the month on the mean grozeira CPUE (P = <0.001), but no effect (P = 0.096) on the mean arraieira CPUE. Grozeira CPUE values were highest in December and March, and lowest between May to August

CID: an efficient complexity-invariant distance for time series

The ubiquity of time series data across almost all human endeavors has produced a great interest in time series data mining in the last decade. While dozens of classification algorithms have been applied to time series, recent empirical evidence strongly suggests that simple nearest neighbor classification is exceptionally difficult to beat. The choice of distance measure used by the nearest neighbor algorithm is important, and depends on the invariances required by the domain. For example, motion capture data typically requires invariance to warping, and cardiology data requires invariance to the baseline (the mean value). Similarly, recent work suggests that for time series clustering, the choice of clustering algorithm is much less important than the choice of distance measure used.In this work we make a somewhat surprising claim. There is an invariance that the community seems to have missed, complexity invariance. Intuitively, the problem is that in many domains the different classes may have different complexities, and pairs of complex objects, even those which subjectively may seem very similar to the human eye, tend to be further apart under current distance measures than pairs of simple objects. This fact introduces errors in nearest neighbor classification, where some complex objects may be incorrectly assigned to a simpler class. Similarly, for clustering this effect can introduce errors by “suggesting” to the clustering algorithm that subjectively similar, but complex objects belong in a sparser and larger diameter cluster than is truly warranted.We introduce the first complexity-invariant distance measure for time series, and show that it generally produces significant improvements in classification and clustering accuracy. We further show that this improvement does not compromise efficiency, since we can lower bound the measure and use a modification of triangular inequality, thus making use of most existing indexing and data mining algorithms. We evaluate our ideas with the largest and most comprehensive set of time series mining experiments ever attempted in a single work, and show that complexity-invariant distance measures can produce improvements in classification and clustering in the vast majority of cases.