Using Non-Additive Measure for Optimization-Based Nonlinear Classification

Over the past few decades, numerous optimization-based methods have been proposed for solving the classification problem in data mining. Classic optimization-based methods do not consider attribute interactions toward classification. Thus, a novel learning machine is needed to provide a better understanding on the nature of classification when the interaction among contributions from various attributes cannot be ignored. The interactions can be described by a non-additive measure while the Choquet integral can serve as the mathematical tool to aggregate the values of attributes and the corresponding values of a non-additive measure. As a main part of this research, a new nonlinear classification method with non-additive measures is proposed. Experimental results show that applying non-additive measures on the classic optimization-based models improves the classification robustness and accuracy compared with some popular classification methods. In addition, motivated by well-known Support Vector Machine approach, we transform the primal optimization-based nonlinear classification model with the signed non-additive measure into its dual form by applying Lagrangian optimization theory and Wolfes dual programming theory. As a result, 2n – 1 parameters of the signed non-additive measure can now be approximated with m (number of records) Lagrangian multipliers by applying necessary conditions of the primal classification problem to be optimal. This method of parameter approximation is a breakthrough for solving a non-additive measure practically when there are relatively small number of training cases available (mn-1). Furthermore, the kernel-based learning method engages the nonlinear classifiers to achieve better classification accuracy. The research produces practically deliverable nonlinear models with the non-additive measure for classification problem in data mining when interactions among attributes are considered.

Integrating geometallurgical ball mill throughput predictions into short-term stochastic production scheduling in mining complexes

This article presents a novel approach to integrate a throughput prediction model for the ball mill into short-term stochastic production scheduling in mining complexes.The datasets for the throughput prediction model include penetration rates from blast hole drilling(measurement while drilling),geological domains,material types,rock density,and throughput rates of the operating mill,offering an accessible and cost-effective method compared to other geometallurgical programs.First,the comminution behavior of the orebody was geostatistically simulated by building additive hardness proportions from penetration rates.A regression model was constructed to predict throughput rates as a function of blended rock properties,which are informed by a material tracking approach in the mining complex.Finally,the throughput prediction model was integrated into a stochastic optimization model for short-term production scheduling.This way,common shortfalls of existing geometallurgical throughput prediction models,that typically ignore the non-additive nature of hardness and are not designed to interact with mine production scheduling,are overcome.A case study at the Tropicana Mining Complex shows that throughput can be predicted with an error less than 30 t/h and a correlation coefficient of up to 0.8.By integrating the prediction model and new stochastic components into optimization,the production schedule achieves weekly planned production reliably because scheduled materials match with the predicted performance of the mill.Comparisons to optimization using conventional mill tonnage constraints reveal that expected production shortfalls of up to 7%per period can be mitigated this way.

Impact of tree litter identity, litter diversity and habitat quality on litter decomposition rates in tropical moist evergreen forest

Background:Attempts to restore degraded highlands by tree planting are common in East Africa.However,up till now,little attention has been given to effects of tree species choice on litter decomposition and nutrient recycling.Method:In this study,three indigenous and two exotic tree species were selected for a litter decomposition study.The objective was to identify optimal tree species combinations and tree diversity levels for the restoration of degraded land via enhanced litter turnover.Litterbags were installed in June 2019 into potential restoration sites(disturbed natural forest and forest plantation)and compared to intact natural forest.The tested tree leaf litters included five monospecific litters,ten mixtures of three species and one mixture of five species.Standard green and rooibos tea were used for comparison.A total of 1,033 litters were retrieved for weight loss analysis after one,three,six,and twelve months of incubation.Results:The finding indicates a significant effect of both litter quality and litter diversity on litter decomposition.The nitrogen-fixing native tree Millettia ferruginea showed a comparable decomposition rate as the fast decom-posing green tea.The exotic conifer Cupressus lusitanica and the native recalcitrant Syzygium guineense have even a lower decomposition rate than the slowly decomposing rooibos tea.A significant correlation was observed be-tween litter mass loss and initial leaf litter chemical composition.Moreover,we found positive non-additive ef-fects for litter mixtures including nutrient-rich and negative non-additive effects for litter mixtures including poor leaf litters respectively.Conclusion:These findings suggest that both litter quality and litter diversity play an important role in decom-position processes and therefore in the restoration of degraded tropical moist evergreen forest.

Heterosis, Combining Ability and Gene Action for Yield in Bottle Gourd

The experiment was conducted at the experimental field of Olericulture Division, Horticulture Research Centre (HRC), Bangladesh Agricultural Research Institute (BARI), Gazipur, Bangladesh during the winter season of 2018-2019 to study the genetic architecture of yield in a seven parent half diallel cross of bottle gourd. The values of mean square for GCA (general combining ability) and SCA (specific combining ability) were highly significant which suggested the presence of both additive and non-additive genetic variance in the population. But the higher magnitude of GCA compared to SCA indicated predominance of additive genetic variance. In most of the cases, the cross between poor and poor parents showed positive SCA effect for fruit yield, which indicated the higher yield. The estimates of significant positive better parent heterosis ranged from 6.27 to 49.72 percent. Analysis of genetic components of variation suggested that additive components were more important in the inheritance of fruit yield. This character was observed being controlled by two to three pairs of genes or groups of genes. Narrow sense heritability was 23 percent indicating probability of selection in generations. The graphical analysis also indicated wide genetic diversity among the parents.

Litter mixing significantly affects decomposition in the Hulun Buir meadow steppe of Inner Mongolia, China

Aims We explored the decomposition rates of single-and mixed-species litter,the litter-mixing effect and the effect of component litters in a mixture on decomposition.Methods In a litter bag experiment,shoot litters from two dominant grasses(Leymus chinensis and Stipa baicalensis)and one legume(Melissitus ruthenica)were decomposed separately and as a mixture from May 2010 to September 2011 in the Hulun Buir meadow steppe of Inner Mongolia,China.We separated the litter mixture into its individual component litters(i.e.the different single-species litters)and analyzed the changes in litter mass remaining and litter nitrogen(N)remaining during single-and mixed-species litter decomposition.Important Findings(i)Litter mixing had significant positive effects on litter decomposition.The litter-mixing effect was strongest for the mixture of S.baicalensis and L.chinensis litters,followed by the mixture of S.baicalensis and M.ruthenica litters.(ii)Single-species component litters decomposed faster in the mixtures than separately(positive effect),but these effects were not significant for legume species M.ruthenica litter.Relative to single-species litter decomposition,the decomposition rates of the two grass(S.baicalensis and L.chinensis)litters significantly increased when they were mixed with each other or with M.ruthenica litter.(iii)For each species litter type,the percentage of litter N remaining during decomposition(NR)differed between the single-species litter and mixed litter treatments.The NR of S.baicalensis litter was higher when it was decomposed in the mixture than in isolation.However,the NR of L.chinensis litter was lowest in its mixture with M.ruthenica among the treatments.Regardless of its decomposition in the mixture or in isolation,the NR of M.ruthenica litter varied little among treatments.There was a significant positive relationship between the NR and percentage of initial litter mass remaining in both the single litter and mixed litter treatments.These results suggest that N transfer may happen among component litters in mixture and further affect the decomposition.

Dynamic Patterns of Gene Expression Additivity and Regulatory Variation throughout Maize Development

Gene expression variation is a key component underlying phenotypic variation and heterosis. Transcriptome profiling was performed on 23 different tissues or developmental stages of two maize inbreds, B73 and Mo17, as well as their hybrid. The obtained large-scale datasets provided opportunities to monitor the developmental dynamics of differential expression, additivity for gene expression, and regulatory variation. The transcriptome can be divided into .30 000 genes that are expressed in at least one tissue of one in bred and an additional ~10 000 “silent” genes that are not expressed in any tissue of any genotype, 90% of which are non-syntenic relative to other grasses. Many (.74%) of the expressed genes exhibit differential expression in at least one tissue. However, the majority of genes with differential expression do not exhibit consistent differential expression in different tissues. These genes often exhibit tissue-specific differential expression with equivalent expression in other tissues, and in many cases they switch the directionality of differential expression in different tissues. This suggests widespread variation for tissue-specific regulation of gene expression between the two maize inbreds B73 and Mo17. Nearly 5000 genes are expressed in only one parent in at least one tissue (single parent expression) and 97% of these genes are expressed at mid-parent levels or higher in the hybrid, providing extensive opportunities for hybrid complementation in heterosis. In general, additive expression patterns are much more common than non-additive patterns, and this trend is more pronounced for genes with strong differential expression or single pare nt expressi on. There is relatively little evidence for non-additive expression patterns that are maintained in multiple tissues. The analysis of allele-specific expression allowed classification of cis. and trans-regulatory variation. Genes with c/s-regulatory variation often exhibit additive expression and tend to have more consistent regulatory variation throughout development. In contrast, genes with trans-reguiatory variation are enriched for non-additive patterns and often show tissue-specific differential expression. Taken together, this study provides a deeper understatiding of regulatory variation and the degree of additive gene expression throughout maize development. The dynamic nature of differential expression, additivity, and regulatory variation imply abundant variability for tissue-specific regulatory mechanisms and suggest that connections between transcriptome and phenome will require expression data from multiple tissues.

A general information theoretical proof for the second law of thermodynamics

It is shown that the conservation and the non-additivity of the information, together with the additivity of the entropy, make the entropy increase in an isolated system. The collapse of the entangled quantum state offers an example of the information non-additivity. Nevertheless, the non-additivity of information is also true in other fields in which the interaction information is important. Examples are classical statistical mechanics, social statistics and financial processes. The second law of thermodynamics is thus proven in its most general form. It is exactly true not only in quantum and classical physics but also in other processes in which the information is conservative and non-additive.