Twitter, one of the biggest and most popular microblogging Websites, has evolved into a powerful communication platform which allows millions of active users to generate huge volume of microposts and queries on a daily basis. To accommodate effective categorization and easy search, users are allowed to make use of hashtags, keywords or phrases prefixed by hash character, to categorize and summarize their posts. However, valid hashtags are not restricted and thus are created in a free and heterogeneous style, increasing difficulty of the task of tweet categorization. In this paper, we propose a low-rank weighted matrix factorization based method to recommend hashtags to the users solely based on their hashtag usage history and independent from their tweets' contents. We confirm using two-sample t-test that users are more likely to adopt new hashtags similar to the ones they have previously adopted. In particular, we formulate the problem of hashtag recommendation into an optimization problem and incorporate hashtag correlation weight matrix into it to account for the similarity between different hashtags. We finally leverage widely used matrix factorization from recommender systems to solve the optimization problem by capturing the latent factors of users and hashtags. Empirical experiments demonstrate that our method is capable to properly recommend hashtags.
Message passing between components of a distributed physical system is non-instantaneous and contributes to determine the time scales of the emerging collective dynamics like an effective inertia. In biological neuron networks this inertia is due in part to local synaptic filtering of exchanged spikes, and in part to the distribution of the axonal transmission delays. How differently these two kinds of inertia affect the network dynamics is an open issue not yet addressed due to the difficulties in dealing with the non-Markovian nature of synaptic transmission. Here, we develop a mean-field dimensional reduction yielding to an effective Markovian dynamics of the population density of the neuronal membrane potential, valid under the hypothesis of small fluctuations of the synaptic current. The resulting theory allows us to prove the formal equivalence between local and distributed inertia, holding for any synaptic time scale, integrate-and-fire neuron model, spike emission regimes and for different network states even when the neuron number is finite.
Whole genome duplication (WGD) is one of the most important events in the molecular evolution of organisms. In fish species, a WGD is considered to have occurred in the ancestral lineage of teleosts. Recent comprehensive ortholog comparisons among teleost genomes have provided useful data and insights into the fate of redundant genes generated by WGD. Based on these data, a mathematical model is proposed to explain the evolutionary scenario of genes after WGD. The model is parameterized taking into account an equilibrium between i) rapid loss of either of the duplicate genes and ii) moderate functional differentiation of each of duplicate genes, both of which are followed by slow gene loss under purifying selection. This model predicts that, in the teleost lineage, a maximum of about 3000 gene pairs may have differentiated functionally during 90 million years after WGD. Thus, the present study provides a possibility that the whole impact of WGD can be quantitatively assessed according to the model parameters, before details of genomic structural changes or functional differentiation are investigated. If the equilibrium model is valid not only for teleosts but also for other lineages that have undergone WGDs, correlations between the assessment indices and evolutionarily significant events, such as the diversification of species or the occurrence of novel phenotypes, could be tested and compared among those lineages.
Recently, Eklund et al. (2016) analyzed clustering methods in standard FMRI packages: AFNI (which we maintain), FSL, and SPM [1]. They claimed: 1) false positive rates (FPRs) in traditional approaches are greatly inflated, questioning the validity of "countless published fMRI studies"; 2) nonparametric methods produce valid, but slightly conservative, FPRs; 3) a common flawed assumption is that the spatial autocorrelation function (ACF) of FMRI noise is Gaussian-shaped; and 4) a 15-year-old bug in AFNI's 3dClustSim significantly contributed to producing "particularly high" FPRs compared to other software. We repeated simulations from [1] (Beijing-Zang data [2], see [3]), and comment on each point briefly.
Recently described stochastic models of protein evolution have demonstrated that the inclusion of structural information in addition to amino acid sequences leads to a more reliable estimation of evolutionary parameters. We present a generative, evolutionary model of protein structure and sequence that is valid on a local length scale. The model concerns the local dependencies between sequence and structure evolution in a pair of homologous proteins. The evolutionary trajectory between the two structures in the protein pair is treated as a random walk in dihedral angle space, which is modelled using a novel angular diffusion process on the two-dimensional torus. Coupling sequence and structure evolution in our model allows for modelling both "smooth" conformational changes and "catastrophic" conformational jumps, conditioned on the amino acid changes. The model has interpretable parameters and is comparatively more realistic than previous stochastic models, providing new insights into the relationship between sequence and structure evolution. For example, using the trained model we were able to identify an apparent sequence-structure evolutionary motif present in a large number of homologous protein pairs. The generative nature of our model enables us to evaluate its validity and its ability to simulate aspects of protein evolution conditioned on an amino acid sequence, a related amino acid sequence, a related structure or any combination thereof.
We consider a non-Gaussian option pricing model, into which the underlying log-price is assumed to be driven by an $\alpha$-stable distribution. We remove the a priori divergence of the model by introducing a Mellin regularization for the L\'evy propagator. Using distributional and $\mathbb{C}^n$ tools, we derive an analytic closed formula for the option price, valid for any stability $\alpha\in]1,2]$ and any asymmetry. This formula is very efficient and recovers previous cases (Black-Scholes, Carr-Wu); we calibrate the formula on market datas, make numerical tests, and discuss its many interesting properties.
In phylogenetic analysis, for non-molecular data, particularly morphology, parsimony optimization is the most commonly employed approach. In the past and present application of the parsimony principle, extra step numbers have been added across different characters without proper justification. This practice, however, has caused the impacts of characters to be inflated or deflated without a valid reason. To resolve this methodological deficiency, here I present a corrected parsimony criterion for reconstructing phylogenies. In essence, character rather than step is the most fundamental unit. Accordingly, the most parsimonious tree should maximize the sum or average of the phylogenetic signals, quantified by retention index, contributed by each character. Assigning proper weights to characters is a separate task that requires information other than the intra-character step number changing range.
STATCHECK is an R algorithm designed to scan papers automatically for inconsistencies between test statistics and their associated p values (Nuijten et al., 2016). The goal of this comment is to point out an important and well-documented flaw in this busily applied algorithm: It cannot handle corrected p values. As a result, statistical tests applying appropriate corrections to the p value (e.g., for multiple tests, post-hoc tests, violations of assumptions, etc.) are likely to be flagged as reporting inconsistent statistics, whereas papers omitting necessary corrections are certified as correct. The STATCHECK algorithm is thus valid for only a subset of scientific papers, and conclusions about the quality or integrity of statistical reports should never be based solely on this program.
Functional neuroimaging studies have lead to understanding the brain as a collection of spatially segregated functional networks. It is thought that each of these networks is in turn composed of a set of distinct sub-regions that together support each network's function. Considering the sub-regions to be an essential part of the brain's functional architecture, several strategies have been put forward that aim at identifying the functional sub-units of the brain by means of functional parcellations. Current parcellation strategies typically employ a bottom-up strategy, creating a parcellation by clustering smaller units. We propose a novel top-down parcellation strategy, using time courses of instantaneous connectivity to subdivide an initial region of interest into sub-regions. We use split-half reproducibility to choose the optimal number of sub-regions. We apply our Instantaneous Connectivity Parcellation (ICP) strategy on high-quality resting-state FMRI data, and demonstrate the ability to generate parcellations for thalamus, entorhinal cortex, motor cortex, and subcortex including brainstem and striatum. We evaluate the subdivisions against available cytoarchitecture maps to show that the our parcellation strategy recovers biologically valid subdivisions that adhere to known cytoarchitectural features.
The conventional wisdom of mean-variance (MV) portfolio theory asserts that the nature of the relationship between risk and diversification is a decreasing asymptotic function, with the asymptote approximating the level of portfolio systematic risk or undiversifiable risk. This literature assumes that investors hold an equally-weighted or a MV portfolio and quantify portfolio diversification using portfolio size. However, the equally-weighted portfolio and portfolio size are MV optimal if and only if asset returns distribution is exchangeable or investors have no useful information about asset expected return and risk. Moreover, the whole of literature, absolutely all of it, focuses only on risky assets, ignoring the role of the risk free asset in the efficient diversification. Therefore, it becomes interesting and important to answer this question: how valid is this conventional wisdom when investors have full information about asset expected return and risk and asset returns distribution is not exchangeable in both the case where the risk free rate is available or not? Unfortunately, this question have never been addressed in the current literature. This paper fills the gap.
We investigate the dynamics of large-scale interacting neural populations, composed of conductance based, spiking model neurons with modifiable synaptic connection strengths, which are possibly also subjected to external noisy currents. The network dynamics is controlled by a set of neural population probability distributions ($\mathrm{PPD}$)which are constructed along the same lines as in the Klimontovich approach to the kinetic theory of plasmas. An exact non-closed, nonlinear, system of integro-partial differential equations is derived for the $\mathrm{PPD}$s. As is customary, a closing procedure leads to a mean field limit. The equations we have obtained are of the same type as those which have been recently derived using rigorous techniques of probability theory. The numerical solutions of these so called McKean-Vlasov-Fokker-Planck equations, which are only valid in the limit of infinite size networks, actually shows that the statistical measures as obtained from $\mathrm{PPD}$s are in good agreement with those obtained through direct integration of the stochastic dynamical system for large but finite size networks. Although numerical solutions have been obtained in the case of Fitzhugh-Nagumo model neurons, the theory can be readily applied to systems of Hodgkin-Huxley type model neurons of arbitrary dimension.
Optimization of fluid transport in the slime mold Physarum polycephalum has been the subject of several modeling efforts in recent literature. Existing models assume that the tube adaptation mechanism in P. polycephalum's tubular network is controlled by the sheer amount of fluid flow through the tubes. We put forward the hypothesis that the controlling variable may instead be the flow's pressure gradient along the tube. We carry out the stability analysis of such a revised mathematical model for a parallel-edge network, proving that the revised model supports the global flow-optimizing behavior of the slime mold for a substantially wider class of response functions compared to previous models. Simulations also suggest that the same conclusion may be valid for arbitrary network topologies.
This work is a systematic study of discrete Markov chains that are used to describe the evolution of a two-types population. Motivated by results valid for the well-known Moran (M) and Wright-Fisher (WF) processes, we define a general class of Markov chains models which we term the Kimura class. It comprises the majority of the models used in population genetics, and we show that many well-known results valid for M and WF processes are still valid in this class. In all Kimura processes, a mutant gene will either fixate or become extinct, and we present a necessary and sufficient condition for such processes to have the probability of fixation strictly increasing in the initial frequency of mutants. This condition implies that there are WF processes with decreasing fixation probability --- in contradistinction to M processes which always have strictly increasing fixation probability. As a by-product, we show that an increasing fixation probability defines uniquely an M or WF process which realises it, and that any fixation probability with no state having trivial fixation can be realised by at least some WF process. These results are extended to a subclass of processes that are suitable for describing time-inhomogeneous dynamics. We also discuss the traditional identification of frequency dependent fitnesses and pay-offs, extensively used in evolutionary game theory, the role of weak selection when the population is finite, and the relations between jumps in evolutionary processes and frequency dependent fitnesses.
On the basis of experimental data and mathematical equations in the literature, we remodel the ionic dynamics of smooth muscle cells (SMCs) as an eigensystem formulation, which is valid for investigating finite variations of variables from the equilibrium like in common experimental operations. This algorithm provides an alternate viewpoint from frequency-domain analysis and enables one to probe functionalities of SMC's rhythm by means of a resonance-related mechanism. Numerical results show three types of calcium oscillations of SMCs in mesenteric arterioles: spontaneous calcium oscillation, agonist-dependent calcium oscillation, and agonist-dependent calcium spike. For simple single and double SMCs, we demonstrate properties of synchronization among complex signals related to calcium oscillations, and show different correlation relations between calcium and voltage signals for various synchronization and resonance conditions. For practical cell clusters, our analyses indicate that the rhythm of SMCs could (1) benefit enhancements of signal communications among remote cells, (2) respond to a significant calcium peaking against transient stimulations for triggering globally-oscillating modes, and (3) characterize the globally-oscillating modes via frog-leap (non-molecular-diffusion) calcium waves across inhomogeneous SMCs.
In multivariate pattern analysis of neuroimaging data, 'second-level' inference is often performed by entering classification accuracies into a $t$-test vs chance level across subjects. We argue that while the random-effects analysis implemented by the $t$-test does provide population inference if applied to activation differences, it fails to do so in the case of classification accuracy or other 'information-like' measures, because the true value of such measures can never be below chance level. This constraint changes the meaning of the population-level null hypothesis being tested, which becomes equivalent to the global null hypothesis that there is no effect in any subject in the population. Consequently, rejecting it only allows to infer that there are some subjects in which there is an information effect, but not that it generalizes, rendering it effectively equivalent to fixed-effects analysis. This statement is supported by theoretical arguments as well as simulations. We review possible alternative approaches to population inference for information-based imaging, converging on the idea that it should not target the mean, but the prevalence of the effect in the population. One method to do so, 'permutation-based information prevalence inference using the minimum statistic', is described in detail and applied to empirical data.
Surface electromyography (s-EMG) sensors are a promising way to control upper-limb prostheses. However a training session is necessary in order to set up the controller that will make s-EMG based movement possible. All data recorded during the training session are used by a machine learning algorithm to make a posture classification, that will allow the controller to distinguish each posture. The aim of this study is to investigate if it's possible to make a posture classification which can remain valid over time. The next step will be the study of how it varies depending on the amount of information submitted to it during the training session in view of real life everyday use of the upper-limb prosthesis.
Purpose - To test major Web search engines on their performance on navigational queries, i.e. searches for homepages. Design/methodology/approach - 100 real user queries are posed to six search engines (Google, Yahoo, MSN, Ask, Seekport, and Exalead). Users described the desired pages, and the results position of these is recorded. Measured success N and mean reciprocal rank are calculated. Findings - Performance of the major search engines Google, Yahoo, and MSN is best, with around 90 percent of queries answered correctly. Ask and Exalead perform worse but receive good scores as well. Research limitations/implications - All queries were in German, and the German-language interfaces of the search engines were used. Therefore, the results are only valid for German queries. Practical implications - When designing a search engine to compete with the major search engines, care should be taken on the performance on navigational queries. Users can be influenced easily in their quality ratings of search engines based on this performance. Originality/value - This study systematically compares the major search engines on navigational queries and compares the findings with studies on the retrieval effectiveness of the engines on informational queries. Paper type - research paper
This paper investigates the composition of search engine results pages. We define what elements the most popular web search engines use on their results pages (e.g., organic results, advertisements, shortcuts) and to which degree they are used for popular vs. rare queries. Therefore, we send 500 queries of both types to the major search engines Google, Yahoo, Live.com and Ask. We count how often the different elements are used by the individual engines. In total, our study is based on 42,758 elements. Findings include that search engines use quite different approaches to results pages composition and therefore, the user gets to see quite different results sets depending on the search engine and search query used. Organic results still play the major role in the results pages, but different shortcuts are of some importance, too. Regarding the frequency of certain host within the results sets, we find that all search engines show Wikipedia results quite often, while other hosts shown depend on the search engine used. Both Google and Yahoo prefer results from their own offerings (such as YouTube or Yahoo Answers). Since we used the .com interfaces of the search engines, results may not be valid for other country-specific interfaces.
Operational risk models commonly employ maximum likelihood estimation (MLE) to fit loss data to heavy-tailed distributions. Yet several desirable properties of MLE (e.g. asymptotic normality) are generally valid only for large sample-sizes, a situation rarely encountered in operational risk. In this paper, we study how asymptotic normality does--or does not--hold for common severity distributions in operational risk models. We then apply these results to evaluate errors caused by failure of asymptotic normality in constructing confidence intervals around the MLE fitted parameters.
Bayesian hierarchical models are a methodology for aggregation and synthesis of data from heterogeneous settings, used widely in statistics and other disciplines. I apply this framework to the evidence from 7 randomized experiments of expanding access to microcredit to assess the general impact of the intervention on household outcomes and the heterogeneity in this impact across sites. The results suggest that the effect of microcredit is likely to be positive but small relative to control group average levels, and the possibility of a negative impact cannot be ruled out. By contrast, common meta-analytic methods that pool all the data without assessing the heterogeneity misleadingly produce "statistically significant" results in 2 of the 6 household outcomes. Standard pooling metrics for the studies indicate on average 60% pooling on the treatment effects, suggesting that the site-specific effects are reasonably externally valid, and thus informative for each other and for the general case. The cross-study heterogeneity is almost entirely generated by heterogeneous effects for the 27% households who previously operated businesses before microcredit expansion, although this group is likely to see much larger impacts overall. A Ridge regression procedure to assess the correlations between site-specific covariates and treatment effects indicates that the remaining heterogeneity is strongly correlated with differences in economic variables, but not with differences in study design protocols. The average interest rate and the average loan size have the strongest correlation with the treatment effects, and both are negative.