Selection Informative Units for Extractive Summarization

We studied outlier document filtering (ODF) for extractive sentence summarization. Our results are superior compared to the average of the participant systems’ using DUC 2006. Furthermore, we add extractive paragraph summarization to the same system. It is surprising that the results are nearly the same for ROUGE metrics. Although extractive paragraph summarization has a better performance for precision, extractive sentence summarization has a slightly better performance on the recall and F-Score which is the harmonic mean of recall and precision. The ODF is successful for both extractive sentence and paragraph summarization. The similarity metric (match percent) suggested in the article prevents the domination of longer sentences/paragraphs on shorter sentences/paragraphs in selection. As a result, the ODF provides the flexibility of paragraph extraction instead of sentence extraction for simplicity and readability and less work load. DOI: http://dx.doi.org/10.5755/j01.itc.43.4.7010

The effects of electrical vestibular stimulation upon movement and perception suggest two evoked sensations: head roll and inter-aural linear acceleration. The head roll vector causes walking subjects to turn in a direction dependent on head pitch, requiring generation of torque around a vertical axis. Here the effect of vestibular stimulation upon vertical torque (T(z)) was investigated during quiet stance. With the head tilted forward, square-wave stimuli applied to the mastoid processes evoked a polarity-specific T(z) response accompanied by trunk yaw. Stochastic vestibular stimulation (SVS) was used to investigate the effect of head pitch with greater precision; the SVS–T(z) cross-correlation displayed a modulation pattern consistent with the head roll vector and this was also reflected by changes in coherence at 2–3 Hz. However, a separate response at 7–8 Hz was unaffected by head pitch. Head translation (rather than rotation) had no effect upon this high frequency response either, suggesting it is not caused by a sense of body rotation induced by an inter-aural acceleration vector offset from the body. Instead, high coherence between medio-lateral shear force and T(z) at the same frequency range suggests it is caused by mechanical coupling to evoked medio-lateral sway. Consistent with this explanation, the 7–8 Hz response was attenuated by 90 deg head roll or yaw, both of which uncouple the inter-aural axis from the medio-lateral sway axis. These results demonstrate two vertical torque responses to electrical vestibular stimulation in standing subjects. The high frequency response can be attributed to mechanical coupling to evoked medio-lateral sway. The low frequency response is consistent with a reaction to a sensation of head roll, and provides a novel method for investigating proprioceptive-vestibular interactions during stance.

Text summarization is becoming an indispensable solution for dealing with the exponential growth of textual and unstructured information in digital format. In this paper, an unsupervised method for extractive multi-document summarization is presented. This method combines the use of a semantic graph for representing textual contents and identify the most relevant topics with the processing of several sentences features applying a fuzzy logic perspective. A fuzzy aggregation operator is applied in the sentences relevance assessment process as a contribution to the multi-document summarization process. The method was evaluated with the Spanish and English texts collection of MultiLing 2015. The obtained results were measured through ROUGE metrics and compared with those obtained by other solutions reported from MultiLing2015.

Random Indexing and Modified Random Indexing based approach for extractive text summarization

Extractive summarization of clinical trial descriptions

In this paper the attenuation of Lamb waves in three-layer joints, including two elastic plates bonded together by a viscoelastic adhesive layer, is investigated in high and low frequencies using global matrix method. Theoretical boundary value problem in three-layer adhesive joints with perfect bond and traction-free boundary conditions on their outer surfaces is solved to find a combination of frequencies and modes with lowest attenuation. Characteristic equation is derived by applying continuity and boundary conditions in three-layer joints using global matrix method. Attenuation and phase velocity dispersion curves in high and low frequencies are obtained with numerical solution of this equation by a computer code for a three-layer joint, including an aluminum repair patch bonded to the aircraft aluminum skin by a layer of viscoelastic epoxy adhesive. To validate the numerical solution results of characteristic equation, wave structure curves are plotted for a special mode in two different frequencies in the adhesive joint. Also, attenuated displacement amplitude curves in a round-trip of Lamb wave along the adhesive joint are plotted to investigate the reduction of the signal amplitude generated by a single transducer. The purpose of present paper is to find a combination of frequencies and modes with minimum attenuation in high and low frequencies. These frequencies and modes are recognizable by transducers in inspections with Lamb waves because of low attenuation level.

Inductively coupled plasmais widely used in semiconductor and display process because of its desirable characteristics such as high plasma density, simple structure and independently controllable ion energy. The driving frequency is a significant parameter that generates and maintains the plasma. However, the effects of different driving frequencies on the radial distribution of the plasma parameters are hardly investigated. So a large area cylindrical inductively coupled plasma source driven separately by 2 MHz and 13.56 MHz is investigated. In order to perform a comprehensive investigation about the effect of driving frequency, the radially resolved measurements of electron density, electron temperature and density of metastable state atoms for the argon discharge are systematically analyzed by Langmuir double probe and optical emission spectroscopy at various power values and gas pressures. It is found that input power values at high frequency (13.56 MHz) and low frequency (2 MHz) have different effects on plasma parameters. When discharge is driven at high frequency, the electron density increases obviously with the increase of power. However, when discharge is driven at low frequency, the electron temperature increases evidently with the increase of power. This can be explained by calculating the skin depths in high and low frequency discharge. When the discharge is driven at high frequency, the induced electromagnetic field is higher than that at low frequency, and the single electron obtains more energy. It is easier to ionize, so the energy is mainly used to increase the electron density. When the discharge is driven at low frequency, the skin layer is thicker, the number of heated electrons is larger, and the average energy of electrons is increased, so the energy is mainly used to raise the electron temperature. At a gas pressure of 10 Pa, the electron density shows a ‘convex’ distribution and increases with the increase of input power for both the high-frequency and low-frequency discharge. While the distributions of electron temperature are obviously different. When the discharge is driven at high frequency, the electron temperature is relatively flat in the center of the chamber and slightly increases on the edge. When the discharge is driven at low frequency, the electron temperature gradually decreases along the radial position. This is due to the one-step ionization in the high-frequency discharge and the two-step ionization in the low-frequency discharge. In order to prove that the low-frequency discharge is dominated by two-step ionization, the spectral intensities of the argon plasma under the same discharge conditions are diagnosed by optical emission spectroscopy. The number density of metastable states is calculated by the branch ratio method. The results are consistent with the analyses. At a gas pressure of 100 Pa, the electron density increases and then decreases with the increase of radial distance, and the overall distribution shows a " saddle shape” for high frequency and also for low frequency discharge. Although the uniformity of electron density improves with the gas pressure, the uniformity at low frequency is better than that at high frequency. The reason can be attributed to the fact that the skin layer of low frequency is thicker and the heating area is wider.

Distortion-product otoacoustic emissions (DPOAEs) are repeatable over time at lower frequencies (<or=8 kHz). The purpose of this study was to examine the repeatability of DPOAEs measured with high-frequency stimuli in normal-hearing adult subjects. It was hypothesized that high-frequency DPOAEs would be repeatable over a minimum of 4 wk. DPOAEs were measured in 25 subjects (14 female and 11 male) with normal behavioral thresholds, normal middle ear function, and present acoustic reflexes at 1 kHz evoked by contralateral stimulation. Behavioral thresholds were measured through 16 kHz, using Békèsy tracking. Each subject attended four trials, in which a complete set of data was collected. The four trials were separated by at least one, but no more than 2 wk. At each trial, two different DPOAE paradigms were completed. Frequency and ratio sweeps were used to measure DPOAE levels and calculate DPOAE group delay, respectively. Frequency sweeps were measured with varied stimulus level conditions (L1/L2 = 60/45, 60/50, 70/55, 70/60 dB SPL) and an f2/f1 of 1.2 at discrete f2 frequencies between 2 and 16 kHz. Ratio sweeps were obtained at f2 frequencies of 2, 4, 6, 8, 10, 12, 14, and 16 kHz, with L1 = 60 and L2 = 45 dB SPL, and the ratio (f2/f1) was varied from 1.05 to 1.3. Repeated-measures analysis of variance was performed. Greater variability was found at the higher frequencies (>8 kHz) for DPOAE level measurements. The average DPOAE level differences-between-trials for the higher and lower frequencies for the four different stimulus level conditions was 5.15 (SD = 4.40 dB) and 2.80 (SD = 2.70 dB) dB, respectively. Individual subject analysis revealed that high-frequency DPOAE levels varied no more than 10 dB for 87.5 and 83.1% of young adult subjects for the 70/55 and 60/50 dB SPL stimulus level conditions, respectively. For low frequencies, repeated DPOAE level variations were within +/-10 dB for 98.4 and 96% of young adult subjects for the 70/55 and 60/50 dB SPL stimulus level conditions, respectively. For DPOAE group delay, greater variability was noted at lower frequencies (<or=8 kHz) and at 10 kHz. The average group-delay differences-between-trials for the higher (10 to 16 kHz) and lower (<or=8 kHz) frequencies was 0.22 (SD = 0.20 msec) and 0.28 (SD = 0.24 msec) msec, respectively. Individual subject analysis revealed that DPOAE group-delay values varied no more than 0.70 msec for the majority of young adult subjects (96.5 and 93.5% at high and low frequencies, respectively). Even though the DPOAE level data obtained at frequencies greater than 8 kHz were more variable than at low frequencies, the higher frequencies were found to be repeatable for both paradigms tested. These results encourage the exploration of high-frequency DPOAE measures to be used as an objective test for monitoring ototoxicity in humans. Testing subjects receiving ototoxic therapies is a necessary step in determining if monitoring high-frequency DPOAEs will successfully predict ototoxic effects.

Many patients with severe obstructive sleep apnea (OSA) do not complain of excessive daytime sleepiness (EDS), possibly due to increased sympathetic nervous activity (SNA) and accompanying heightened alertness. We hypothesized that in patients with OSA, those without subjective EDS (Epworth Sleepiness Scale, ESS score < 11) would have higher very low frequency (VLF) heart rate variability (HRV) during sleep, reflecting greater sympathetic heart rate modulation than patients with an ESS score ≥ 11. Patients with severe OSA (AHI ≥ 30: 26 with and 65 without heart failure) were divided into those with and without EDS. Heart rate (HR) signals were acquired in stage 2 sleep during periods of recurrent apneas and hypopneas and submitted to coarse graining spectral analysis, which extracts harmonic, neurally mediated contributions to HRV from total spectral power. Because the apnea-hyperpnea cycle entrains muscle SNA at VLF (0 to 0.04 Hz), VLF power was our principal between-group comparison. Subjects without EDS had higher harmonic VLF power (944 ± 839 vs 447 ± 461 msec(2), p = 0.003) than those with EDS, irrespective of the presence or absence of heart failure (1218 ± 944 vs 426 ± 299 msec(2), p = 0.043, and 1029 ± 873 vs 503 ± 533 msec(2), p = 0.003, respectively). ESS scores correlated inversely with VLF power in all (r = -0.294, p = 0.005) and in heart failure subjects (r = -0.468, p = 0.016). Patients with severe OSA but without EDS have higher VLF-HRV than those with EDS. This finding suggests that patients with severe OSA but without EDS have greater sympathetic modulation of HRV than those with EDS that may reflect elevated adrenergically mediated alertness. Taranto Montemurro L; Floras JS; Picton P; Kasai T; Alshaer H; Gabriel JM; Bradley TD. Relationship of heart rate variability to sleepiness in patients with obstructive sleep apnea with and without heart failure.

In order to figure out the dynamics response of vibrator baseplate and why baseplate performs poorly at high frequencies, a finite element model of vibrator-ground is developed to analyze acceleration, deformation and decoupling of the baseplate at low and high frequencies. Results show that the acceleration at low frequencies is in accordance with the function of hydraulic pressure. On the contrary, the acceleration at high frequencies is irregular and fluctuant. Within the bandwidth, the maximum deformation locates in the central area, and the baseplate deformation in long-axis direction is greater than that in minor-axis direction. Compared to low frequencies, dynamics response at high frequencies suffers response delay, and the vibrator structure is very sensitive to asymmetry, resulting in warping and waveform deformation. Baseplate experiences decoupling at both low and high frequencies, however, the decoupling at high frequencies occurs more frequently, which leads to more serious distortion. It is also found that symmetrical structure is helpful for reducing negative deformation. In conclusion, high fluctuation of acceleration, response delay and harmonic caused by complicated deformation and decoupling distortion reduce the quality of output signal and block the bandwidth towards high frequencies.

&amp;lt;p&amp;gt;Climate change has a significant role in increasing extreme precipitation, including the intensity, frequency, and magnitude of events due to increases in atmospheric moisture and climate variability. This means that future increases in floods due to climate change must be considered in the construction of flood defenses, as well as the planning of new infrastructure and hydraulic structures. Previous approaches for stress testing the design of flood defenses have relied on the scenario neutral approach and the use of harmonic functions to represent changes in the seasonality and mean of precipitation. Such approaches may inadequately account for changes in extreme precipitation, especially in runoff dominated catchments. Here, we adapt the scenario neutral approach by integrating a discrete wavelet transform (DWT) to develop the flood response surface. Such an approach allows evaluation of flood sensitivity to high and low frequency components of precipitation. Using 39 catchments in Ireland, we examine the sensitivity of flooding (QT20) to changes in low and high frequency precipitation and air temperature. A sensitivity domain of 525 extreme precipitation scenarios is applied by combining 21 low frequency and 25 high frequency sets of precipitation and air temperature changes, with short duration frequency incorporated in each harmonic wavelet function. Clustering and discriminant analysis are used to create a typology of catchment sensitivity based on generated response surfaces, the mean of annual maximum precipitation, and the mean of annual maximum flows. Results allow characterization of catchment sensitivity in gauged and ungauged locations and the integration of a wider spectrum of precipitation changes when assessing sensitivity allowances for climate change. &amp;amp;#160;&amp;lt;/p&amp;gt;&amp;lt;p&amp;gt;&amp;amp;#160;&amp;lt;/p&amp;gt;

In a dual frequency capacitive sheath, a high frequency uniform sheath motion is coupled with a low frequency Child law sheath motion. For current-driven high and low frequency sheaths, the high frequency sheath motion generates most of the ohmic and stochastic heating of the discharge electrons. The low frequency motion, in addition to its primary purpose of establishing the ion bombarding energy, also increases the heating by widening the sheath width and transporting the oscillating electrons to regions of lower plasma density, and hence higher sheath velocity. In this work, we show that for voltage-driven high and low frequency sheaths, increasing the low frequency voltage reduces the heating, due to the reduced high frequency current that flows through the sheath under voltage-driven conditions. We determine the dependence of the heating on various parameters and compare the results with the current-driven case. Particle-in-cell simulations are used to confirm this result. Discharges generally employ a matching network to maximize the power transmitted to the plasma. We obtain analytic expressions for the effect of the low frequency source under matched conditions and, again, find that the low frequency source reduces the heating.

(2010). Linguistic Assessment Tools for the Digisonic® Dual Electric-Acoustic Speech Processor. Cochlear Implants International: Vol. 11, Proceedings of the 9th European Symposium on Paediatric Cochlear Implantation, Warsaw, 2009, pp. 306-311.

The dorsal aspect target strength functions of ten species of fish collected by bottom trawl were measured at two frequencies. The average TS were calculated with respect to fish tilt angle distribution, which is assumed to have a mean value of -5 degrees and a standard deviation of 15 degrees. Both maximum TS and average TS were normalized with the fish length (L[cm]) squared and the 2/3 power of body weight (W[g]), namely, TS=20logL+A, TS=20/3log W+B.The results obtained were as follows:(1) The target strength is generally higher at low frequency (25kHz) than at high frequency (100kHz). But for two species (Chub mackerel and Kandari), neither maximum nor average TS differ between low and high frequencies.(2) The target strength of Silver pomfret, a bladderless fish, is clearly lower than that of bladder fish at two frequencies.(3) The normalized TS (A or B) were calculated for the maximum TS (Amax or Bmax) and average TS (Aavg or Bavg). The results are: Amax = -63 - -59 dB, Aavg = -68 - -64 dB at low frequency; Amax = -65 - -59 dB, Aavg = -72 - -65 dB at high frequency, and Bmax = -52 - -45 dB, Bavg = -57 - -50 dB at low frequency; Bmax = -54 - -45 dB, Bavg = -60 - -52 dB at high frequency.

Objective To investigate the effect of different flickering light frequencies for inducing myopia in C57BL/6 mice.Methods It was a animal experimental study.Ninety mice at age 28 days with similar refractions in both eyes were randomly assigned to five groups of 15 mice each.The control group was exposed to white non-flickering light.Three groups were exposed to white flickering lights at a high frequency,middle frequency,and low frequency at 10,5,and 2 Hz,respectively.A form-deprivation group consisted of 30 mice that underwent monocular form-deprivation with a semitransparent hemispherical thin plastic shell covering the right eyes,and was exposed to a non-flickering white light.Refraction and axial length were measured by murine-specific eccentric infrared photorefraction and A-scan ultrasonography at baseline and after two weeks of exposure.The data were analyzed by one way ANOVA and LSD t methods.Results ①The change in refraction and axial length in mice:the baselines of the refractions of the control group,high frequency,middle frequency and low frequency flickering light groups and the form-deprivation group were 2.627±0.494 D,2.597±0.626 D,2.552±0.558 D,2.617±0.568 D,2.590±0.657 D,respectively.The baselines of the axial length for the corresponding groups were 2.805±0.029 mm,2.803±0.026 mm,2.804±0.029 mm,2.806±0.026 mm,2.804±0.025 mm,respectively.There was no significant difference in refraction or axial length among the five groups (P＞0.05).After 2 weeks of exposure,the refraction of the control group,the high frequency,middle frequency,low frequency flickering light groups and the form-deprivation group were 3.720-±0.555 D,-1.710±0.632 D,-2.020±0.705 D,-2.612±0.686 D,-2.873±0.670 D,respectively.The axial lengths for the corresponding groups were 2.924±0.022 mm,2.995±0.024 mm,3.000±-0.020 mm,3.017±0.029 mm,3.026±0.016 mm,respectively.There was a significant difference in refraction and axial length between the control group and the other groups (P＜0.01).However,the induced changes in refraction and axial length were greater in the low frequency flickering light group and the form-deprivation group compared to the high and middle frequency flickering light groups.Conclusion B6 mice could be induced to form myopia by flickering light at all frequencies we used,and the low frequency of 2 Hz was the most effective in inducing myopia that is equivalent to form-deprivation myopia. Key words: Flickering light; frequency ; Myopic model ; Mouse