Term Extraction Method Research Articles

A Combined Term Extraction Method for the Problem of Monitoring Thematic Discussions in Social Media

A Combined Term Extraction Method for the Problem of Monitoring Thematic Discussions in Social Media

Box–Behnken Design to Optimize Standardized Mangiferin-Rich Mango Peel Extract from Agro-Industrial Waste Product

The light mango or “Ma-Muang Bao” (Mangifere indica L. var.) is a native mango species originating from Malaysia and southern Thailand. The whole Ma-Muang Bao fruit, except peels, is popular as both a raw and ripe fruit for consumption, as well as being used in various processed food products. This study aims to transform the peel of a specific mango variety, which is a byproduct of the agro-industrial sweet, pickled mangoes industry, into a valuable bioactive ingredient for healthcare products. This is achieved through the establishment of a standardized mangiferin-rich mango peel extract (SMPE). Employing the Box–Behnken design (BBD) within the framework of response surface methodology (RSM), an optimal microwave-assisted extraction procedure was developed. A total of 27 experiments, each with four independent variables, including solvent ratio, extraction power, extraction time, and ethanol (EtOH) ratio, were conducted to optimize the extraction method in terms of mangiferin content and extraction yield. The optimized extraction conditions encompassed a solvent ratio of 120 mL EtOH/100 g sample, an extraction power of 450 W, an extraction time of approximately 4.3 min, and an EtOH ratio of 69.44% (EtOH in water). Small-scale extractions were carried out using the following specified parameters: solvent ratio of 120 mL, extraction power of 450 W, extraction time of 4 min, and EtOH ratio of 70% EtOH. These extractions yielded an extract with a mangiferin content of 27.24 ± 2.05 mg/g and an extraction yield of 3.71 ± 0.17% w/w. Notably, these outcomes were better from the mangiferin content of 19.62 mg/g and a yield of fresh peel of 5.61% estimated through BBD analysis. Furthermore, a pilot-scale extraction was performed using 7 kg of fresh mango peel and 70% EtOH (8.4 L) for 4 min, resulting in an extract with a mangiferin content of 51.85 ± 0.35 mg/g and a fresh peel yield of 4.35% w/w. This method emerges as the most suitable for mango peel extraction and forms the basis of the SMPE. The results from biological activities highlight the potential use of SMPE as the active ingredient for cosmeceutical or healthcare products for wound-healing and skin-brightening agents. Additionally, the knowledge from this study presents an alternative approach to various plant sources and sustainable extraction methods for the herbal extract industry.

Soft prompt enhanced joint learning for cross-domain aspect-based sentiment analysis

Aspect term extraction is a fundamental task in fine-grained sentiment analysis, aiming to detect customer's opinion targets from reviews about products or services. The traditional supervised models have achieved promising results with annotated datasets. However, their performance dramatically decreases in cross-domain aspect term extraction tasks. Existing cross-domain transfer learning methods face two common limitations: (1) these works directly inject linguistic features into language models, making it challenging to transfer linguistic knowledge to the target domain; (2) they rely on the fixed predefined prompts, which is time-consuming to construct the prompts for all potential aspect term spans. To address the limitations, we propose a soft prompt-based joint learning method for cross-domain aspect term extraction in this paper. Specifically, by incorporating external linguistic features, the proposed method learns domain-invariant representations between source and target domains via multiple objectives, which bridges the gap between domains with varied distributions of aspect terms. Furthermore, the proposed method interpolates a set of transferable soft prompts consisting of multiple learnable vectors that are beneficial to detect aspect terms in the target domain. Extensive experiments are conducted on two groups of datasets and the experimental results demonstrate the effectiveness of the proposed method for cross-domain aspect terms extraction.

Research and application of intelligent hyperspectral analysis technology for Chinese materia medica

With the development of imaging technology and artificial intelligence, hyperspectral imaging technology provides a fast, non-destructive, intelligent, and precise new method for the analysis of Chinese materia medica(CMM). This paper summarized the methods and applications of hyperspectral imaging technology combined with intelligent analysis technology in the field of CMM in recent years, focusing on the acquisition and preprocessing of hyperspectral data, intelligent analysis methods of hyperspectral data, and practical cases of these technologies in the field of CMM. Hyperspectral data of CMM can provide spectral information with nanometer-level resolution and rich spatial texture information simultaneously. This paper summarized the acquisition process, including black-and-white board calibration and region-of-interest extraction, and preprocessing methods including smoothing, differentiation, scale-space, and scattering correction. The feature extraction methods in terms of spectral, spatial, color, and texture were briefly described, and common modeling methods were summarized. Finally, this paper reviewed the research cases of the application of the above methods to the fields of CMM, such as authenticity identification, origin tracing, variety recognition, year identification, sulfur fumigation degree determination, and quantitative measurement.

Effect of Different Extraction Methods on the Total Phenolics of Sugar Cane Products.

The health benefits of sugar cane products are attributed to certain antioxidant compounds in plant materials. The presence of antioxidants in plant materials depends on the extraction method in terms of yield and the number of phenolic compounds identified. This study was carried out to evaluate the performance of the three extraction methods, which were selected from previous studies to show the effect of the extraction method on the content of antioxidant compounds in different types of sugar. This study also evaluates the potential of different sugar extracts in anti-diabetic activity based on in vitro assays (α-glucosidase and α-amylase). The results showed that sugar cane extracted with acidified ethanol (1.6 M HCl in 60% ethanol) was the best condition to extract a high yield of phenolic acids compared to other methods. Among the three types of sugar, less refined sugar (LRS) showed the highest yield of phenolic compounds, 57.72 µg/g, compared to brown sugar (BS) and refined sugar (RS) sugar, which were at 42.19 µg/g and 22.06 µg/g, respectively. Whereas, among the sugar cane derivatives, LRS showed minor and BS moderate inhibition towards α-amylase and α-glucosidase activity compared to white sugar (RS). Thus, it is suggested that sugar cane extracted with acidified ethanol (1.6 M HCl in 60% ethanol) is the optimum experimental condition for antioxidant content determination and provides a basis for further exploitation of the health-beneficial resources of the sugarcane products.

Emerging Issues in Mapping Urban Impervious Surfaces Using High-Resolution Remote Sensing Images

Urban impervious surface (UIS) is a key parameter in climate change, environmental change, and sustainability. UIS extraction has been evolving rapidly in the past decades. However, high-resolution impervious surface mapping is a long-term need. There is an urgent requirement for impervious surface mapping from high-resolution remote sensing imagery. In this paper, we compare current extraction methods in terms of extraction units and extraction models and summarize their strengths and limitations. We discuss the challenges in impervious surface estimation from high spatial resolution remote sensing imagery in terms of selection of spatial resolution, spectral band, and extraction method. The uncertainties caused by clouds and snow, shadows, and vegetation occlusion are also analyzed. Automated sample labeling and remote sensing domain knowledge are the main directions in impervious surface extraction using deep learning methods. We should also focus on using continuous time series of high-resolution imagery and multi-source satellite imagery for dynamic monitoring of impervious surfaces.

Salt Solubilization Coupled with Membrane Filtration-Impact on the Structure/Function of Chickpea Compared to Pea Protein.

The demand for pulse proteins as alternatives to soy protein has been steeply increasing over the past decade. However, the relatively inferior functionality compared to soy protein is hindering the expanded use of pulse proteins, namely pea and chickpea protein, in various applications. Harsh extraction and processing conditions adversely impact the functional performance of pea and chickpea protein. Therefore, a mild protein extraction method involving salt extraction coupled with ultrafiltration (SE-UF) was evaluated for the production of chickpea protein isolate (ChPI). The produced ChPI was compared to pea protein isolate (PPI) produced following the same extraction method in terms of functionality and feasibility of scaling. Scaled-up (SU) ChPI and PPI were produced under industrially relevant settings and evaluated in comparison to commercial pea, soy, and chickpea protein ingredients. Controlled scaled-up production of the isolates resulted in mild changes in protein structural characteristics and comparable or improved functional properties. Partial denaturation, modest polymerization, and increased surface hydrophobicity were observed in SU ChPI and PPI compared to the benchtop counterparts. The unique structural characteristics of SU ChPI, including its ratio of surface hydrophobicity and charge, contributed to superior solubility at both a neutral and acidic pH compared to both commercial soy protein and pea protein isolates (cSPI and cPPI) and significantly outperformed cPPI in terms of gel strength. These findings demonstrated both the promising scalability of SE-UF and the potential of ChPI as a functional plant protein ingredient.

Conventional and Enzyme-Assisted Extraction of Total Flavonoids from Rotten Onion (Allium cepa L.)

One of the richest sources of dietary flavonoids is the bulb of the onion (Allium cepa L.), which greatly contributes to the daily intake of flavonoids. This research aimed to optimize the yield of Total flavonoids from Onion Solid Waste (OSW). A comparison was drawn between Conventional and Enzyme Assisted extraction methods in terms of the yield of total flavonoids present in OSW extract. The results showed that the highest TFC were obtained at Time/Temp 60 °C for 180 minutes and Pectinase Conc./Time 2g for 120 minutes for Conventional and Enzyme assisted extraction respectively. The yield of TFC by both methods showed that the yield for both was higher in case of enzyme assisted extraction for T13, i.e., 0.472 mg/g respectively than that for conventional extraction for T4 i.e., 0.352 mg/g.

A Hierarchical Information Extraction Method for Large-Scale Centralized Photovoltaic Power Plants Based on Multi-Source Remote Sensing Images

In the context of global sustainable development, solar energy is very widely used. The installed capacity of photovoltaic panels in countries around the world, especially in China, is increasing steadily and rapidly. In order to obtain accurate information about photovoltaic panels and provide data support for the macro-control of the photovoltaic industry, this paper proposed a hierarchical information extraction method, including positioning information and shape information, and carried out photovoltaic panel distribution mapping. This method is suitable for large-scale centralized photovoltaic power plants based on multi-source satellite remote sensing images. This experiment takes the three northwest provinces of China as the research area. First, a deep learning scene classification model, the EfficientNet-B5 model, is used to locate the photovoltaic power plants on 16-m spatial resolution images. This step obtains the area that contains or may contain photovoltaic panels, greatly reducing the study area with an accuracy of 99.97%. Second, a deep learning semantic segmentation model, the U2-Net model, is used to precisely locate photovoltaic panels on 2-m spatial resolution images. This step achieves the exact extraction results of the photovoltaic panels from the area obtained in the previous step, with an accuracy of 97.686%. This paper verifies the superiority of a hierarchical information extraction method in terms of accuracy and efficiency through comparative experiments with DeepLabV3+, U-Net, SegNet, and FCN8s. This meaningful work identified 180 centralized photovoltaic power plants in the study area. Additionally, this method makes full use of the characteristics of different remote sensing data sources. This method can be applied to the rapid extraction of global photovoltaic panels.

Consistency of Feature Importance Algorithms for Interpretable EEG Abnormality Detection.

Recent advances in machine learning show great potential for automatic detection of abnormalities in electroencephalography (EEG). While simple and interpretable models combined with expert-comprehensible input features offer full control of the decision making process, these methods commonly lag behind complex deep learning and feature extraction methods in terms of performance. Here we study a feasibility of a bridging solution, where deep learning is combined with interpretable input and an algorithm computing the importance of particular EEG features in the decision process. We built a convolutional neural network with multi-channel EEG frequency bands as input and investigated four different methods for feature importance attribution: Layer-wise Relevance Propagation (LRP), DeepLIFT, Integrated Gradients (IG) and Guided GradCAM. Our analysis showed consistency between the first three methods, and deviating attributions of the fourth method, suggesting the importance of using a package of methods together to ensure the robustness of medical interpretation.

Amoxicillin removal from medical wastewater using an eco-friendly aqueous two-phase extraction system

ABSTRACT Antibiotics are a common medical waste that can cause antibiotic resistance in humans and harm aquatic life. One of the most effective approaches for treating medical wastewater is the aqueous two-phase extraction process (ATPS). ATPS is considered an ecologically benign technology. Nine ATPS chemical systems were developed to extract Amoxicillin from simulated hospital wastewater using various salts and organic solvents. The criteria that determine the most efficient system were explored in order to maximize the separation process. After assessing the performance of the considered systems, amoxicillin extraction was carried out utilizing an ATPS comprised of PEG6000 as an extractant and K2HPO4 as a separating agent. Complete amoxicillin removal was attained at room temperature and a mixing speed of 500 rpm for 540 mins. The extraction efficiency was found to increase with temperature, salt concentration, and the volume ratio of the solvent (PEG6000). Based on the experimental design methodology, the optimum amoxicillin removal efficiency of 96.4% was attained at a temperature of 44.3°C, 42.6 wt.% of salt, and a 0.5 volume fraction ratio of PEG6000. The stagewise application of the process was also performed within five separation stages. This approach outperforms traditional extraction methods in terms of efficacy and duration.

Chinese-Uyghur Bilingual Lexicon Extraction Based on Weak Supervision

Bilingual lexicon extraction is useful, especially for low-resource languages that can leverage from high-resource languages. The Uyghur language is a derivative language, and its language resources are scarce and noisy. Moreover, it is difficult to find a bilingual resource to utilize the linguistic knowledge of other large resource languages, such as Chinese or English. There is little related research on unsupervised extraction for the Chinese-Uyghur languages, and the existing methods mainly focus on term extraction methods based on translated parallel corpora. Accordingly, unsupervised knowledge extraction methods are effective, especially for the low-resource languages. This paper proposes a method to extract a Chinese-Uyghur bilingual dictionary by combining the inter-word relationship matrix mapped by the neural network cross-language word embedding vector. A seed dictionary is used as a weak supervision signal. A small Chinese-Uyghur parallel data resource is used to map the multilingual word vectors into a unified vector space. As the word-particles of these two languages are not well-coordinated, stems are used as the main linguistic particles. The strong inter-word semantic relationship of word vectors is used to associate Chinese-Uyghur semantic information. Two retrieval indicators, such as nearest neighbor retrieval and cross-domain similarity local scaling, are used to calculate similarity to extract bilingual dictionaries. The experimental results show that the accuracy of the Chinese-Uyghur bilingual dictionary extraction method proposed in this paper is improved to 65.06%. This method helps to improve Chinese-Uyghur machine translation, automatic knowledge extraction, and multilingual translations.

Multicomponent, multiwavelength benchmarks for source- and filament-extraction methods

Modern multiwavelength observations of star-forming regions that reveal complex, highly structured molecular clouds require adequate extraction methods that provide both complete detection of the structures and their accurate measurements. The omnipresence of filamentary structures and their physical connection to prestellar cores make it necessary to use methods that are able to disentangle and extract both sources and filaments. It is fundamentally important to test all extraction methods on standard benchmarks to compare their detection and measurement qualities and fully understand their capabilities before their scientific applications. A recent publication described getsf, the new method for source and filament extraction that employs the separation of the structural components, a successor to getsources, getfilaments, and getimages (collectively referred to as getold). This new paper describes a detailed benchmarking of both getsf and getold using two multicomponent, multiwavelength benchmarks resembling the Herschel observations of the nearby star-forming regions. Each benchmark consists of simulated images at six Herschel wavelengths and one additional derived surface density image with a 13″ resolution. The structural components of the benchmark images include a background cloud, a dense filament, hundreds of starless and protostellar cores, and instrumental noise. Five variants of benchmark images of different complexity are derived from the two benchmarks and are used to perform the source and filament extractions with getsf and getold. A formalism for evaluating source detection and measurement qualities is presented, allowing quantitative comparisons of different extraction methods in terms of their completeness, reliability, and goodness, as well as the detection and measurement accuracies and the overall quality. A detailed analysis of the benchmarking results shows that getsf has better source and filament extraction qualities than getold and that the best choice of the images for source detection with getsf is the high-resolution surface density, alone or with the other available Herschel images. The benchmarks explored in this paper provide the standard tests for calibrating existing and future source- and filament-extraction methods to choose the best tool for astrophysical studies.

Utilising heterogeneous language resources for term extraction in maritime domains

Abstract The development of terminologies for domains where these are lacking is a time-consuming and costly task. This article takes a methodological perspective and addresses a general methodological question: how can we, with limited funding, utilise to a maximal degree, existing language resources to create a terminology at a relatively low cost? Although an important player in the maritime industries for many centuries, Norway has not prioritised the systematic development of an official maritime terminology. The article therefore focuses specifically on efforts to develop a national resource for maritime domains. The article describes efforts to create a corpus of popular science and a parallel corpus of technical texts. Six different term extraction methods are applied. These include corpus-based statistical analyses of frequency, collocation and keyness, as well as bilingual term extraction. Finally, the pros and cons of each method are evaluated by means of a cost-benefit analysis.

Detection of fabric defects with intertwined frame vector feature extraction

In textile industry, defect detection plays a vital role in diagnosing and improving the process. However, fabric defect detection is a process performed by the human eye, as methods that provide fast and sufficient performance are not yet widely used. In this paper, a new feature extraction method for defect detection is presented that is faster and provides higher accuracy. The developed method is basically a feature extraction process from a window taken from the image. Intertwined frames are defined in the matrix which form the window. A center of gravity is defined for each frame. For each frame, a frame vector is drawn from the center pixel to the center of gravity. The results of various vector functions among the frame vectors are used as features. These features are used in classification algorithms for defect detection. The AITEX dataset was used, and the proposed method was compared with the traditional texture feature extraction methods in terms of performance and time. It has been shown that the proposed method the proposed method works 55% faster than the fastest algorithm and provides at least 1.8% more accurate results. In this way, the method can be used on low-level devices.

Method for Automatic Term Extraction from Scientific Articles Based on Weak Supervision

We propose a method for scientific terms extraction from the texts in Russian based on weakly supervised learning. This approach doesn't require a large amount of hand-labeled data. To implement this method we collected a list of terms in a semi-automatic way and then annotated texts of scientific articles with these terms. These texts we used to train a model. Then we used predictions of this model on another part of the text collection to extend the train set. The second model was trained on both text collections: annotated with a dictionary and by a second model. Obtained results showed that giving additional data, annotated even in an automatic way, improves the quality of scientific terms extraction.

EX-Action: Automatically Extracting Threat Actions from Cyber Threat Intelligence Report Based on Multimodal Learning

With the increasing complexity of network attacks, an active defense based on intelligence sharing becomes crucial. There is an important issue in intelligence analysis that automatically extracts threat actions from cyber threat intelligence (CTI) reports. To address this problem, we propose EX-Action, a framework for extracting threat actions from CTI reports. EX-Action finds threat actions by employing the natural language processing (NLP) technology and identifies actions by a multimodal learning algorithm. At the same time, a metric is used to evaluate the information completeness of the extracted action obtained by EX-Action. By the experiment on the CTI reports that consisted of sentences with complex structure, the experimental result indicates that EX-Action can achieve better performance than two state-of-the-art action extraction methods in terms of accuracy, recall, precision, and F1-score.

Updated Lithological Map in the Forest Zone of the Centre, South and East Regions of Cameroon Using Multilayer Perceptron Neural Network and Landsat Images

The Multilayer Perceptron Neural Network (MLPNN) induction technique has been successfully applied to a variety of machine learning tasks, including the extraction and classification of image features. However, not much has been done in the application of MLPNN on images obtained by remote sensing. In this article, two automatic classification systems used in image feature extraction and classification from remote sensing data are presented. The first is a combination of two models: a MLPNN induction technique, integrated under ENVI (Environment for Visualizing Images) platform for classification, and a pre-processing model including dark subtraction for the calibration of the image, the Principal Components Analysis (PCA) for band selections and Independent Components Analysis (ICA) as blind source separator for feature extraction of the Landsat image. The second classification system is a MLPNN induction technique based on the Keras platform. In this case, there was no need for pre-processing model. Experimental results show the two classification systems to outperform other typical feature extraction and classification methods in terms of accuracy for some lithological classes including Granite1 class with the highest class accuracies of 96.69% and 92.69% for the first and second classification system respectively. Meanwhile, the two classification systems perform almost equally with the overall accuracies of 53.01% and 49.98% for the first and second models respectively though the keras model has the advantage of not integrating the pre-processing model, hence increasing its efficiency. The application of these two systems to the study area resulted in the generation of an updated geological mapping with six lithological classes detected including the Gneiss, the Micaschist, the Schist and three versions of Granites (Granite1, Granite2 and Granite3).

RETRACTED: Network data mining based on artificial intelligence inference engine

Demand for rapid development of computerized reporting of the contract, the framework program tireless identity of a generalization of electronic records in the form of slogans, is a matter of major concern. In this paper, the term extraction method based on the use of text mining equipment business Word Stat and Compare-Suite Pro, devised a method to imaginative deployment of a large number of Expert System (ES) techniques. These processes include ES categories, including: Rule-based framework, information-based, case-based, wise experts, and based on the main frame fuzzy, object placement strategy, Depending on the position and the study in the article tile sets most repeated words, theories and buzzwords single ES strategy to select keywords. Build a framework for a main sequence induction query based, the extraction process is based on the slogan Policy. Catchword frame is sorted based on the weight of the sentence According to isolate the slogan, the planned a deduction to in data mining. The product will see the slogans of different ES process articles and all other system all the different articles using age-dependent rule set. First, the creation of an induction search to be adjusted to accommodate articles remaining articles produced, and then approve the use of the remaining items. The results show a very high accuracy rate approval. Artificial intelligence is useful computerized assets, can be integrated with information science and research, in order to provide a more feasible, faster way to break down information. Finally, the estimate can be further improved by limiting the policy test equipment for future business plans and receive orders.

Application-based Evaluation of Automatic Terminology Extraction

The aim of this paper is to evaluate performance of several automatic term extraction methods which can be easily utilized by translators themselves. The experiments are conducted on German newspaper articles in the domain of politics on the topic of Brexit. However, they can be easily replicated on any other topic or language as long as it is supported by all three tools used. The paper first provides an extensive introduction into the field of automatic terminology extraction. Next, selected terminology extraction methods are assessed using precision with respect to the gold standard compiled on the same corpus. Moreover, the corpus has been completely annotated to allow for the calculation of recall. The effects of using five cut-off points are examined in order to find an optimal value which should be used in translation practice.