Extracting lexical and phrasal paraphrases: a review of the literature

Recent advances in natural language processing have increased the popularity of paraphrase extraction. Most of the attention, however, has been focused on the extraction methods only without taking the resource factor into the consideration. Unknowingly, there is a strong relationship between them and the resource factor also plays an equally important role in paraphrase extraction. In addition, almost all of the previous studies have been focused on corpus-based methods that extract paraphrases from corpora based solely on syntactic similarity. Despite the popularity of corpus-based methods, a considerable amount of research has consistently shown that these methods are vulnerable to several types of erroneous paraphrases. For these reasons, it is necessary to evaluate whether the trend is moving in a positive direction. This paper reviews the major research on paraphrase extraction methods in detail. It begins by exploring the definition of paraphrase from different perspectives to provide a better understanding of the concept of paraphrase extraction. It then studies the characteristics and potential uses of different types of paraphrase resources. After that, it divides paraphrase extraction methods into four main categories: heuristic-based, knowledge-based, corpus-based and hybrid-based and summarizes their strengths and weaknesses. This paper concludes with some potential open research issues for future directions.     

An integrated semantic-based approach in concept based video retrieval

Multimedia content has been growing quickly and video retrieval is regarded as one of the most famous issues in multimedia research. In order to retrieve a desirable video, users express their needs in terms of queries. Queries can be on object, motion, texture, color, audio, etc. Low-level representations of video are different from the higher level concepts which a user associates with video. Therefore, query based on semantics is more realistic and tangible for end user. Comprehending the semantics of query has opened a new insight in video retrieval and bridging the semantic gap. However, the problem is that the video needs to be manually annotated in order to support queries expressed in terms of semantic concepts. Annotating semantic concepts which appear in video shots is a challenging and time-consuming task. Moreover, it is not possible to provide annotation for every concept in the real world. In this study, an integrated semantic-based approach for similarity computation is proposed with respect to enhance the retrieval effectiveness in concept-based video retrieval. The proposed method is based on the integration of knowledge-based and corpus-based semantic word similarity measures in order to retrieve video shots for concepts whose annotations are not available for the system. The TRECVID 2005 dataset is used for evaluation purpose, and the results of applying proposed method are then compared against the individual knowledge-based and corpus-based semantic word similarity measures which were utilized in previous studies in the same domain. The superiority of integrated similarity method is shown and evaluated in terms of Mean Average Precision (MAP).     

Batch RNAi selector: a standalone program to predict specific siRNA candidates in batches with enhanced sensitivity

RNA interference (RNAi) is a popular and effective method for silencing gene expression. siRNAs should be gene-specific and effective to achieve specific and potent gene silencing. However, most currently available siRNA design programs are web-based programs that either require each sequence be submitted individually, making large-batch analyses difficult to conduct, or only provide limited options for searching off-target candidates (e.g. NCBI-BLAST). We have developed a stand-alone, enhanced RNAi design program that overcomes these shortcomings. We have implemented WU-BLAST, FASTA and SSEARCH homology searches for siRNA candidates to improve gene specific siRNA selection and to identity siRNA candidates that could lead to off-target gene silencing. We also included many new features such as siRNA score calculation and calculation of siRNA internal stability to help select highly potent siRNAs. This program is freely available for academic and commercial use (), and can be installed and run on any Linux machine. Our program automates the search for siRNAs and the resulting data files including a list of siRNA primers with scores and database search results for each siRNA candidate are stored locally for easy retrieval and inspection when needed.     

Naturally-occurring folates in foods: Method development and analysis using liquid chromatography–tandem mass spectrometry (LC–MS/MS)

An accurate method for detecting and quantifying both synthetic (folic acid) and naturally-occurring folates in foods is described. A system capable of analysing the five most commonly occurring folates (pteroylglutamic acid, 5-methyltetrahydrofolate, tetrahydrofolate, 10-formylfolate and 5-formyltetrahydrofolate) in 20 min using liquid chromatography–tandem mass spectrometry (LC–MS/MS) was developed. Quantification of folates was performed using ¹³C labelled internal standards. This paper outlines the development of a comparatively fast LC–MS/MS method, method validation using commercially available folate standards and establishment of the method’s suitability for quantification using selected reaction monitoring (SRM) mass spectrometry. The application of the system was verified by analysing several certified reference materials and comparing results with certified values as determined by microbiological assay. LC–MS/MS promises to be an ideal tool for the quantitative analysis of folates in food.     

Integration of machine learning and first principles models

Model building and parameter estimation are traditional concepts widely used in chemical, biological, metallurgical, and manufacturing industries. Early modeling methodologies focused on mathematically capturing the process knowledge and domain expertise of the modeler. The models thus developed are termed first principles models (or white-box models). Over time, computational power became cheaper, and massive amounts of data became available for modeling. This led to the development of cutting edge machine learning models (black-box models) and artificial intelligence (AI) techniques. Hybrid models (gray-box models) are a combination of first principles and machine learning models. The development of hybrid models has captured the attention of researchers as this combines the best of both modeling paradigms. Recent attention to this field stems from the interest in explainable AI (XAI), a critical requirement as AI systems become more pervasive. This work aims at identifying and categorizing various hybrid models available in the literature that integrate machine-learning models with different forms of domain knowledge. Benefits such as enhanced predictive power, extrapolation capabilities, and other advantages of combining the two approaches are summarized. The goal of this article is to consolidate the published corpus in the area of hybrid modeling and develop a comprehensive framework to understand the various techniques presented. This framework can further be used as the foundation to explore rational associations between several models.     

Enzyme-Assisted Process for Production of Superior Quality Vanilla Extracts from Green Vanilla Pods Using Tea Leaf Enzymes

Vanilla planifolia Andrews is a perennial tropical vine and is an orchid grown for its pleasant flavor. There is an increasing trend world over for using natural flavors. Vanilla being an important food flavoring ingredient, the demand for natural vanilla extract is increasing. Hence, the aim of the present study was to prepare vanilla extract from green beans without going through the elaborate and time-consuming conventional curing process. Vanilla beans after size reduction were mixed in a suitable proportion with tea leaf enzyme extract (TLEE) and incubated to facilitate action of enzymes on vanilla flavor precursors. The beans mix was squeezed, and the filtrate was treated with ethanol to extract the vanilla flavor. TLEE-treated extracts had higher vanillin content (4.2%) compared to Viscozyme extract (2.4%). Also, it had higher intensity of vanilla flavor, sweet, and floral notes. Further, electronic nose analysis confirmed the discrimination between extracts. It was concluded that the use of TLEE is very much useful to obtain higher yield of vanilla extract and superior quality vanilla flavor, which avoids the traditional laborious and time-consuming curing process.     

Chiral Photodetector Based on GaAsN

The detection of light helicity is key for various applications, from drug production to optical communications. However, the light helicity direct measurement is inherently impossible with conventional photodetectors based on III–V or IV–VI non-chiral semiconductors. The prior polarization analysis by often moving optical elements is necessary before light is sent to the detector. A method is here presented to effectively give the conventional dilute nitride GaAs-based semiconductor epilayer a chiral photoconductivity. The detection scheme relies on the giant spin-dependent recombination of conduction electrons and the accompanying spin polarization of the engineered defects to control the conduction band. As the conduction electron spin polarization is, in turn, intimately linked to the excitation light polarization, the light polarization state and intensity can be determined by a simple conductivity measurement. This approach, removing the need for any optical elements in front of a non-chiral detector, could offer easier integration and miniaturization. This new chiral photodetector could potentially operate in a spectral range from the visible to the infra-red using (InGaAl)AsN alloys or ion-implanted nitrogen-free III–V compounds.     

Breast cancer detection by using associative classifier with rule refinement method based on relevance feedback

Neural Computing and Applications - Computer-aided diagnosis system that uses classification process for an automated detection of breast cancer could provide a second opinion that improves...     

Synthesis, characterization and photocatalytic H2 generation over ternary indium titanate nanoparticles

A low temperature surfactant-free solution-phase method has been successfully developed for the synthesis of ternary In2TiO5, nanoparticles using a solvothermal route. The mechanistic aspects of synthesis of In2TiO5 nanoparticles from precursors, In(acac)3 and Ti(IV) isopropoxide in benzyl alcohol at 220 degrees C under solvothermal conditions, were investigated by GC-MS and 13C{1H} NMR analysis. The N2-BET surface area of the 5-8 nm sized In2TiO5 nanoparticles was found to be 60 m2 g(-1), which decreased with increase in calcination temperature; 38 m2 g(-1) at 800 degrees C; 5 m2 g(-1) at 1200 degrees C. The High resolution transmission electron microscopy (HR-TEM) shows well-developed lattice fringes of the crystalline nanoparticles, and selected area electron diffraction (SAED), pattern was indexed to be orthorhombic In2TiO5. The nanoparticles show better photocatalytic hydrogen generation from water-methanol mixtures over bulk In2TiO5, anatase TiO2 nanoparticles prepared by identical route and commercial TiO2 photocatalyst (Degussa, P25) under UV-visible irradiation (16% UV + 84% visible). Photocatalytic properties as a function of crystallinity and surface area of indium titanate nanoparticles have also been investigated. The high photoactivity obtained is correlated with the electronic and crystal structure of In2TiO5.     

Robust Polyphonic Music Retrieval with N-grams

In this paper we investigate the retrieval performance of monophonic and polyphonic queries made on a polyphonic music database. We extend the n-gram approach for full-music indexing of monophonic music data to polyphonic music using both rhythm and pitch information. We define an experimental framework for a comparative and fault-tolerance study of various n-gramming strategies and encoding levels. For monophonic queries, we focus in particular on query-by-humming systems, and for polyphonic queries on query-by-example. Error models addressed in several studies are surveyed for the fault-tolerance study. Our experiments show that different n-gramming strategies and encoding precision differ widely in their effectiveness. We present the results of our study on a collection of 6366 polyphonic MIDI-encoded music pieces.