基于词条向量和情感分析的新闻广告匹配模型

在新闻网页上配置广告的算法通常以信息关联度为基础,没有考虑新闻和广告之间的情感差异,容易导致错误匹配.针对目前广告匹配算法的缺陷,在关键词生成、情感分析技术的基础上,引入情感差异分析,建立了一种新的支持在线新闻与广告内容、情感相匹配的模型.该模型对在线新闻网站和广告投放商实际应用具有重要的意义.

标准分及其应用

本文通过原始分数转化为标准分数或标准百分的方法的介绍 ,说明了标准分数在运动选项 。

Prediction of Subcellular Localization of Eukaryotic Proteins Using Position-Specific Profiles and Neural Network with Weighted Inputs

Subcellular location is one of the key biological characteristics of proteins. Position-specific profiles （PSP） have been introduced as important characteristics of proteins in this article. In this study, to obtain position-specific profiles, the Position Specific lterative-Basic Local Alignment Search Tool （PSI-BLAST） has been used to search for protein sequences in a database. Position-specific scoring matrices are extracted from the profiles as one class of characteristics. Four-part amino acid compositions and lst-7th order dipeptide compositions have also been calculated as the other two classes of characteristics. Therefore, twelve characteristic vectors are extracted from each of the protein sequences. Next, the characteristic vectors are weighed by a simple weighing function and inputted into a BP neural network predictor named PSP-Weighted Neural Network （PSP-WNN）. The Levenberg-Marquardt algorithm is employed to adjust the weight matrices and thresholds during the network training instead of the error back propagation algorithm. With a jackknife test on the RH2427 dataset, PSP-WNN has achieved a higher overall prediction accuracy of 88.4% rather than the prediction results by the general BP neural network, Markov model, and fuzzy k-nearest neighbors algorithm on this dataset. In addition, the prediction performance of PSP-WNN has been evaluated with a five-fold cross validation test on the PK7579 dataset and the prediction results have been consistently better than those of the previous method on the basis of several support vector machines, using compositions of both amino acids and amino acid pairs. These results indicate that PSP-WNN is a powerful tool for subcellular localization prediction. At the end of the article, influences on prediction accuracy using different weighting proportions among three characteristic vector categories have been discussed. An appropriate proportion is considered by increasing the prediction accuracy.

Experimental Study on the Initial Position Distribution of Taylor Bubbles in Cryogenic Upward Inclined Tubes

An experimental study was carried out to understand the phenomena of the boiling flow of liquid nitrogen in inclined tubes with closed bottom by using the high speed digital camera.The tubes in the experiment are 0.018 m and 0.014 m in inner diameter and 1.0 m in length.The range of the inclination angles is 0-45° from the vertical.The statistical method is employed to analyze the experimental data.The experiment was focused on the effect of the inclination angle on the initial position distribution of Taylor bubbles.The formation criterion of Taylor bubbles was confirmed by analyzing the images of Taylor bubbles.The experimental results show that the initial position of Taylor bubble increased first,and then decreased with the increasing inclination angle,with the maximum at 30°.The standard deviation of the initial position of Taylor bubble in tubes was different with different inner diameters.The lognormal shape was fitted to the measured the initial position distributions of Taylor bubbles in the cryogenic tubes.

Reducing Uncertainty in Subdivision Optimization

Design of watertight subdivision inherently involves its optimization with the objective to increase the index ＂A＂ above its minimum required value. In view of a big popularity of probabilistic search methods such as genetic algorithms, this task is intrinsically time consuming. Thus, even when an optimal subdivision layout （i.e. topology） is determined, it can be found that the optimal bulkhead positions can be a great challenge time-wise, often forcing designers to satisfy with suboptimal solutions. The fundamental reason why this happens is that the nature of the optimized function （e.g., index ＂A＂ as a function of bulkhead positions） is unknown and hence it has no effect upon the choice of optimization strategy, which therefore reflects subjective but not factual preferences. In this paper we study the nature of functional dependency between the subdivision index and bulkhead positions, as a simplest case, and indicate pertinent optimization strategies that consequently reduce the optimization time. In our study we use a cruise ship model to demonstrate the application results of our findings.

Combined GPS/GLONASS Data Processing

To obtain the GLONASS satellite position at an epoch other than reference time,the satellite’s equation of motion has to be integrated with broadcasting ephemerides.The iterative detecting and repairing method of cycle slips based on triple difference residuals for combined GPS/GLONASS positioning and the iterative ambiguity resolution approach suitable for combined post processing positioning are discussed systematically.Experiments show that millimeter accuracy can be achieved in short baselines with a few hours’ dual frequency or even single frequency GPS/GLONASS carrier phase observations,and the precision of dual frequency observations is distinctly higher than that of single frequency observations.

Unprecedented Retreat in a 50-Year Observational Record for Petermann Glacier, North Greenland

Petermann Glacier is a marine-terminating outlet glacier that had a 70 km-long floating ice tongue prior to a ～ 270 km2 calving event that was observed from satellite sensors in August 2010,shortening the ice tongue by ～ 27 km.Further,in July 2012,another 10 km was lost through calving.In order to understand these events in perspective,here the authors perform a long-term data analysis of Petermann Glacier calving-front variability and ice velocity for each year in the 1990s-2000s,supplemented by available observations from the previous three decades.Five major （on the order of 100 krm2） calving events are identified,with ～ 153 km2 calved from 1959 to 1961,～ 168 km2 in 1991,～ 71 km2 in 2001,～ 270 km2 in 2010,and ～ 130 km2 in 2012-as well as ～ 31 k m2 calved in 2008.The increased frequency of major calving events in recent years has left the front terminus position retreated nearly 25 km beyond the range of observed in previous decades.In contrast,stable ice-dynamics are suggested from ice-velocity measurements made each year between 1993-2012,which are on average 1063 m yr-1,with limited interannual variability and no significant trend; moreover,there is no apparent relationship between ice-velocity variability and calving events.The degree to which the massive calving events in 2010 and 2012 represent natural episodic variability or a response to atmospheric and/or oceanic changes remains speculative; however,melt-induced weakening of the floating ice tongue in recent years is strongly suggested.

Study on destressing technology for a roadway driven along goaf in a fully mechanized top-coal caving face

Based on the deformation characteristics of the roadways driven along goaf in fully mechanized top coal caving faces, the author considers that it is the key to ensure the stability of surrounding rocks of roadway driven along goaf to control the deformation during the period affected by mining. Considering the characteristics of the roadway layout in fully mechanized top coal caving faces, a technical scheme of destressing is put forward and the destressing effect is analyzed by using the software of Universal Distinct Element Code 3 0(UDEC 3 0).

Constructing confidence intervals of extreme rainfall quantiles using Bayesian,bootstrap,and profile likelihood approaches

Hydrological risk is highly dependent on the occurrence of extreme rainfalls.This fact has led to a wide range of studies on the estimation and uncertainty analysis of the extremes.In most cases,confidence intervals(CIs)are constructed to represent the uncertainty of the estimates.Since the accuracy of CIs depends on the asymptotic normality of the data and is questionable with limited observations in practice,a Bayesian highest posterior density(HPD)interval,bootstrap percentile interval,and profile likelihood(PL)interval have been introduced to analyze the uncertainty that does not depend on the normality assumption.However,comparison studies to investigate their performances in terms of the accuracy and uncertainty of the estimates are scarce.In addition,the strengths,weakness,and conditions necessary for performing each method also must be investigated.Accordingly,in this study,test experiments with simulations from varying parent distributions and different sample sizes were conducted.Then,applications to the annual maximum rainfall(AMR)time series data in South Korea were performed.Five districts with 38-year(1973–2010)AMR observations were fitted by the three aforementioned methods in the application.From both the experimental and application results,the Bayesian method is found to provide the lowest uncertainty of the design level while the PL estimates generally have the highest accuracy but also the largest uncertainty.The bootstrap estimates are usually inferior to the other two methods,but can perform adequately when the distribution model is not heavy-tailed and the sample size is large.The distribution tail behavior and the sample size are clearly found to affect the estimation accuracy and uncertainty.This study presents a comparative result,which can help researchers make decisions in the context of assessing extreme rainfall uncertainties.