Document re-ranking by generality in bio-medical information retrieval

Document ranking is an important process in information retrieval (IR). It presents retrieved documents in an order of their estimated degrees of relevance to query. Traditional document ranking methods are mostly based on the similarity computations between documents and query. In this paper we argue that the similarity-based document ranking is insufficient in some cases. There are two reasons. Firstly it is about the increased information variety. There are far too many different types documents available now for user to search. The second is about the users variety. In many cases user may want to retrieve documents that are not only similar but also general or broad regarding a certain topic. This is particularly the case in some domains such as bio-medical IR. In this paper we propose a novel approach to re-rank the retrieved documents by incorporating the similarity with their generality. By an ontology-based analysis on the semantic cohesion of text, document generality can be quantified. The retrieved documents are then re-ranked by their combined scores of similarity and the closeness of documents’ generality to the query’s. Our experiments have shown an encouraging performance on a large bio-medical document collection, OHSUMED, containing 348,566 medical journal references and 101 test queries.

A web recommendation technique based on probabilistic latent semantic analysis

Web transaction data between Web visitors and Web functionalities usually convey user task-oriented behavior pattern. Mining such type of click-stream data will lead to capture usage pattern information. Nowadays Web usage mining technique has become one of most widely used methods for Web recommendation, which customizes Web content to user-preferred style. Traditional techniques of Web usage mining, such as Web user session or Web page clustering, association rule and frequent navigational path mining can only discover usage pattern explicitly. They, however, cannot reveal the underlying navigational activities and identify the latent relationships that are associated with the patterns among Web users as well as Web pages. In this work, we propose a Web recommendation framework incorporating Web usage mining technique based on Probabilistic Latent Semantic Analysis (PLSA) model. The main advantages of this method are, not only to discover usage-based access pattern, but also to reveal the underlying latent factor as well. With the discovered user access pattern, we then present user more interested content via collaborative recommendation. To validate the effectiveness of proposed approach, we conduct experiments on real world datasets and make comparisons with some existing traditional techniques. The preliminary experimental results demonstrate the usability of the proposed approach.

Genomic and proteomic analysis of synaptic protein expression in the brains of human alcoholics

Alcoholism results in changes in the human brain which reinforce the cycle of craving and dependency, and these changes are manifest in the pattern of expression of mRNA and proteins in key cells and brain areas. Long-term alcohol abuse also results in damage to selected regions of the cortex. We have used cDNA microarrays to show that less than 1% of mRNA transcripts differ signifi cantly between cases and controls in the susceptible area and that the expression profi le of a subset of these transcripts is suffi cient to distinguish alcohol abusers from controls. In addition, we have utilized a 2D gel proteomics based approach to determine the identity of proteins in the superior frontal cortex (SFC) of the human brain that show differential expression in controls and long term alcohol abusers. Overall, 182 proteins differed by the criterion of > 2-fold between case and control samples. Of these, 139 showed signifi cantly lower expression in alcoholics, 35 showed signifi cantly higher expression, and 8 were new or had disappeared. To date 63 proteins have been identifi ed. The expression of one family of proteins, the synucleins, has been further characterized using Real Time PCR and Western Blotting. The expression of alpha-synuclein mRNA was signifi cantly lower in the SFC of alcoholics compared with the same area in controls (P = 0.01) whereas no such difference in expression was found in the motor cortex. The expression of beta- and gamma- synuclein were not signifi cantly different between alcoholics and controls. In contrast, the pattern of alphasynuclein protein expression differs from that of the corresponding RNA transcript. Because of the key role of synaptic proteins in the pathogenesis of alcoholism, we are developing 2-D DIGE based techniques to quantify expression changes in synaptosomes prepared from the SFC of controls and alcoholics.

Noise, regularizers, and unrealizable scenarios in online learning from restricted training sets

We study the dynamics of on-line learning in multilayer neural networks where training examples are sampled with repetition and where the number of examples scales with the number of network weights. The analysis is carried out using the dynamical replica method aimed at obtaining a closed set of coupled equations for a set of macroscopic variables from which both training and generalization errors can be calculated. We focus on scenarios whereby training examples are corrupted by additive Gaussian output noise and regularizers are introduced to improve the network performance. The dependence of the dynamics on the noise level, with and without regularizers, is examined, as well as that of the asymptotic values obtained for both training and generalization errors. We also demonstrate the ability of the method to approximate the learning dynamics in structurally unrealizable scenarios. The theoretical results show good agreement with those obtained by computer simulations.

Examining crystallographic orientation dependence of hardness of silica stishovite

We conducted nanoindentation to explore the hardness and elastic properties of silica stishovite, synthesized at high pressure and quenched to ambient conditions. A total of 10 crystallographic orientations were examined on selected grains with a maximum load of 4 or 20 mN. We observed discontinuity in the load-displacement curve (pop-in) for the [2 5 over(1, -)] and [6 2 over(1, -)] grains subjected to a maximum load of 20 mN. The single-crystal hardness at high plastic deformation is quasi-isotropic with an average of 32 ± 1 GPa, similar to the polycrystalline hardness reported earlier; the theoretical hardness determined from the experiments is about 54 ± 3 GPa. These two hardnesses suggest that stishovite is one of the hardest oxides. The measured indentation moduli are close to the predictions at low load (minor plasticity) but are considerably lower at high load (high plasticity). Both indentation hardness and modulus decrease with increasing plasticity. Our results underscore the necessity of considering the degree of plastic deformation when interpreting hardness and elastic moduli from indentation experiments. © 2007 Elsevier B.V. All rights reserved.

Benefit segmentation of international travelers to Macau, China

The purpose of this study is threefold: (1) to identify the underlying benefits sought by international visitors to Macau, China, which has emerged as a popular gambling destination in Asia; (2) to segment tourists visiting Macau by employing a cluster analysis based on the benefits sought; and (3) to examine any salient differences between the segment groups with regard to their behavioral characteristics, socio-economic characteristics, and demographic profiles. A convenience sample was used to collect data in the Macau International Airport, in the Macau Ferry Terminal, and at the border gate with Mainland China. A total 1,513 useful surveys were retained for data analysis. Cluster analysis discloses four distinct clusters: "convention and business seekers," "family and vacation seekers," "gambling and shopping seekers," and "multi-purpose seekers." Based on the results of our findings, several managerial implications are discussed. © Taylor & Francis Group, LLC.

Single-cell analysis for bioprocessing

Cell population heterogeneity has attracted great interest for understanding the individual cellular performances in their response to external stimuli and in the production of targeted products. Physical characterization of single cells and analysis of dynamic gene expression, synthesized proteins, and cellular metabolites from one single cell are reviewed. Advanced techniques have been developed to achieve high-throughput and ultrahigh resolution or sensitivity. Single cell capture methods are discussed as well. How to make use of cellular heterogeneities for maximizing cellular productivity is still in the infant stage, and control strategies will be formulated after the causes for heterogeneity have been elucidated.

Identifying spatially overlapping local cortical networks with MEG

Recent modelling studies (Hadjipapas et al. [2009]: Neuroimage 44:1290-1303) have shown that it may be possible to distinguish between different neuronal populations on the basis of their macroscopically measured (EEG/MEG) mean field. We set out to test whether the different orientation columns contributing to a signal at a specific cortical location could be identified based on the measured MEG signal. We used 1.5deg square, static, obliquely oriented grating stimuli to generate sustained gamma oscillations in a focal region of primary visual cortex. We then used multivariate classifier methods to predict the orientation (left or right oblique) of the stimuli based purely on the time-series data from this one location. Both the single trial evoked response (0-300 ms) and induced post-transient power spectra (300-2,300 ms, 20-70 Hz band) due to the different stimuli were classifiable significantly above chance in 11/12 and 10/12 datasets respectively. Interestingly, stimulus-specific information is preserved in the sustained part of the gamma oscillation, long after perception has occurred and all neuronal transients have decayed. Importantly, the classification of this induced oscillation was still possible even when the power spectra were rank-transformed showing that the different underlying networks give rise to different characteristic temporal signatures. © 2009 Wiley-Liss, Inc.