Mining sentiments from songs using latent dirichlet allocation

Song-selection and mood are interdependent. If we capture a song’s sentiment, we can determine the mood of the listener, which can serve as a basis for recommendation systems. Songs are generally classified according to genres, which don’t entirely reflect sentiments. Thus, we require an unsupervised scheme to mine them. Sentiments are classified into either two (positive/negative) or multiple (happy/angry/sad/...) classes, depending on the application. We are interested in analyzing the feelings invoked by a song, involving multi-class sentiments. To mine the hidden sentimental structure behind a song, in terms of “topics”, we consider its lyrics and use Latent Dirichlet Allocation (LDA). Each song is a mixture of moods. Topics mined by LDA can represent moods. Thus we get a scheme of collecting similar-mood songs. For validation, we use a dataset of songs containing 6 moods annotated by users of a particular website.

Obtaining single document summaries using latent dirichlet allocation

In this paper, we present a novel approach that makes use of topic models based on Latent Dirichlet allocation(LDA) for generating single document summaries. Our approach is distinguished from other LDA based approaches in that we identify the summary topics which best describe a given document and only extract sentences from those paragraphs within the document which are highly correlated given the summary topics. This ensures that our summaries always highlight the crux of the document without paying any attention to the grammar and the structure of the documents. Finally, we evaluate our summaries on the DUC 2002 Single document summarization data corpus using ROUGE measures. Our summaries had higher ROUGE values and better semantic similarity with the documents than the DUC summaries.

Maximum magnitude estimation considering the regional rupture character

The main objective of the paper is to develop a new method to estimate the maximum magnitude (M (max)) considering the regional rupture character. The proposed method has been explained in detail and examined for both intraplate and active regions. Seismotectonic data has been collected for both the regions, and seismic study area (SSA) map was generated for radii of 150, 300, and 500 km. The regional rupture character was established by considering percentage fault rupture (PFR), which is the ratio of subsurface rupture length (RLD) to total fault length (TFL). PFR is used to arrive RLD and is further used for the estimation of maximum magnitude for each seismic source. Maximum magnitude for both the regions was estimated and compared with the existing methods for determining M (max) values. The proposed method gives similar M (max) value irrespective of SSA radius and seismicity. Further seismicity parameters such as magnitude of completeness (M (c) ), ``a'' and ``aEuro parts per thousand b `` parameters and maximum observed magnitude (M (max) (obs) ) were determined for each SSA and used to estimate M (max) by considering all the existing methods. It is observed from the study that existing deterministic and probabilistic M (max) estimation methods are sensitive to SSA radius, M (c) , a and b parameters and M (max) (obs) values. However, M (max) determined from the proposed method is a function of rupture character instead of the seismicity parameters. It was also observed that intraplate region has less PFR when compared to active seismic region.

Exploiting Chinese character models to improve speech recognition performance

The Chinese language is based on characters which are syllabic in nature. Since languages have syllabotactic rules which govern the construction of syllables and their allowed sequences, Chinese character sequence models can be used as a first level approximation of allowed syllable sequences. N-gram character sequence models were trained on 4.3 billion characters. Characters are used as a first level recognition unit with multiple pronunciations per character. For comparison the CU-HTK Mandarin word based system was used to recognize words which were then converted to character sequences. The character only system error rates for one best recognition were slightly worse than word based character recognition. However combining the two systems using log-linear combination gives better results than either system separately. An equally weighted combination gave consistent CER gains of 0.1-0.2% absolute over the word based standard system. Copyright © 2009 ISCA.

Character of simplified Navier-Stokes (SNS) equations near separation point for two-dimensional laminar flow over a flat plate

It is proved that the simplified Navier-Stokes (SNS) equations presented by Gao Zhi[1], Davis and Golowachof-Kuzbmin-Popof (GKP)[3] are respectively regular and singular near a separation point for a two-dimensional laminar flow over a flat plate. The order of the algebraic singularity of Davis and GKP equation[2,3] near the separation point is indicated. A comparison among the classical boundary layer (CBL) equations, Davis and GKP equations, Gao Zhi equations and the complete Navier-Stokes (NS) equations near the separation point is given.

Orientation preference and fractal character of short fatigue cracks in a weld metal

Short fatigue crack behaviour in a weld metal has been further investigated. The Schmid factor and the fractal dimension of short cracks on iso-stress specimens subjected to reversed bending have been determined and then applied to account for the distribution and orientation characteristics of short fatigue cracks. The result indicates that the orientation preference of short cracks is attributed to the large values of Schmid factor at relevant grains. The Schmid factors of most slip systems, which produced short cracks, are less than or equal to 0.4. Crack length measurements reveal that short crack path, compared to that of long crack, possesses a more stable and relatively larger value of fractal dimension. This is regarded as one of the typical features of short cracks.

Contributions to the Geology and Paleontology of the Canal Zone, Panama, and geologically related areas in central America and the West Indies. Part 11: The biologic character and geologic correlations of the sedimentary formations of Panama in their relation to the geologic history of Central America and the West Indies

The paper presents: 1) biologic summaries for each of the formations for which paleontologic data are available, with brief discussions of the geologic age; 2) geologic correlations of the formations and the distribution of their age-equivalents in Central America, the West Indies, and the southeastern United States; 3) an outline of the paleogeography of middle America. The biologic summaries are based on the paleontologic memoirs in this vol. by Messars. Howe, Berry, Chuchman, Jackson, Canu and Bassler and Pilsbry, Miss Rathbun and myself.

Detección automática de presencia/ausencia de atún en imágenes obtenidas mediante sonar de largo alcance a bordo de buques pesqueros y aplicación de Optical Character Recognition (OCR) para la extracción de parámetros

En esta tesis de máster se presenta una metodología para el análisis automatizado de las señales del sonar de largo alcance y una aplicación basada en la técnica de reconocimiento óptico de Optical Character Recognition, caracteres (OCR). La primera contribución consiste en el análisis de imágenes de sonar mediante técnicas de procesamiento de imágenes. En este proceso, para cada imagen de sonar se extraen y se analizan las regiones medibles, obteniendo para cada región un conjunto de características. Con la ayuda de los expertos, cada región es identi cada en una clase (atún o no-atún). De este modo, mediante el aprendizaje supervisado se genera la base de datos y, a su vez, se obtiene un modelo de clasi cación. La segunda contribución es una aplicación OCR que reconoce y extrae de las capturas de pantalla de imágenes de sonar, los caracteres alfanuméricos correspondientes a los parámetros de situación (velocidad, rumbo, localización GPS) y la confi guración de sonar (ganancias, inclinación, ancho del haz). El objetivo de este proceso es el de maximizar la e ficiencia en la detección de atún en el Golfo de Vizcaya y dar el primer paso hacia el desarrollo de un índice de abundancia de esta especie, el cual esté basado en el procesamiento automático de las imágenes de sonar grabadas a bordo de la ota pesquera durante su actividad pesquera rutinaria.

Meristic variation in Sebastes (Scorpaenidae), with an analysis of character association and bilateral pattern and their significance in species separation

This report presents meristic data for nearly all of the known species of Sebasles. Rudimentary caudal ray counts tend to be higher in more active species. The number of caudal rays supported by the hypurals is consistently 14, whereas the number of branched caudal rays varies between 11 and 13. Vertebral counts and most fin-ray counts tend to be lower in species or populations in warmer latitudes, except for pectoral ray counts which tend to have an opposite geographic pattern. On the basis of the small magnitude of meristic and morphometric differences and the lack of other differences between northern and southern samples of "Sebasles caurinus," Sebaslichlhys vexillaris Jordan and Gilbert is regarded as a junior synonym of Sebasles caurinus Richardson. The patterns of bilateral variation in paired meristics are analyzed and their mechanism discussed. The frequency distribution of pectoral ray counts in their right-left combination is shown to be useful in species separation. No association was found between any combination of two meristic features in any species. The author proposes that intrasample associations between meristic features are evidence of sampling heterogeneity. (PDF file contains 21 pages.)