Document Representation Research Articles

Knowledge Graph-Based Hierarchical Text Semantic Representation

Document representation is the basis of language modeling. Its goal is to turn natural language text that flows into a structured form that can be stored and processed by a computer. The bag-of-words model is used by most of the text-representation methods that are currently available. And yet, they do not consider how phrases are used in the text, which hurts the performance of tasks that use natural language processing later on. Representing the meaning of text by phrases is a promising area of future research, but it is hard to do well because phrases are organized in a hierarchy and mining efficiency is low. In this paper, we put forward a method called hierarchical text semantic representation using the knowledge graph (HTSRKG), which uses syntactic structure features to find hierarchical phrases and knowledge graphs to improve how phrases are evaluated. First, we use CKY and PCFG to build the syntax tree sentence by sentence. Second, we walk through the parse tree using the hierarchical routing process to obtain the mixed phrase semantics in passages. Finally, the introduction of the knowledge graph improves the efficiency of text semantic extraction and the accuracy of text representation. This gives us a solid foundation for tasks involving natural language processing that come after. Extensive testing on actual datasets shows that HTSRKG surpasses baseline approaches with respect to text semantic representation, and the results of a recent benchmarking study support this.

Application of a stand-alone RTI measuring system with an integrated camera in cultural heritage digitisation

The Multi-Light Image Collection (MLIC) data, in most cases processed with the Reflectance Transformation Imaging (RTI) technique, is one of the popular methods of documentation and digital representation of movable finds from an archaeological excavation. The manual (hand-held) approach to collecting this data is technically simple and still popular in the case of archaeology. However, an extensive amount of work during image collection is a strong motivation for building RTI acquisition systems, which are also popular, in most cases, in the shape of domes. The research presented in the article focuses on two aspects of building such a device. First, the study examines the effect of the placement and number of lights on the resulting RTI model for a system based on LED stripe-shaped circuits. Secondly, the camera component was studied.Recent development in the area of low-cost accessories for microcomputers resulted in a few new industrial-grade USB cameras with the possibility to take high-resolution images. Cameras designed for use with microcomputers offer broad control of the focus or exposition parameters and provide the possibility to build RTI domes with the fully integrated camera, therefore, almost complete automatization of the capturing and processing of MLIC data.This paper presents the main assumptions and elements of the dome-shaped measurement system, which enables automatic RTI data acquisition and processing together with the evaluation of light configuration used to build relightable models. Amphoras handles with stamps and ancient pottery were used as an objects to evaluate the quality of the relightable models generated from the images acquired by a programmable miniature Arducam camera. Results of comparison with professional high-quality mirrorless cameras show that the quality of images from miniature cameras can be satisfactory for some archaeological objects.

Improving coordination skills in young football players aged 8-10 years

The assumptions underlying the current direction of research have been determined by both the decreasing level of the football game in the investigated age group and the failure to exploit the impressive skills that a child aged 8-10 years is able to develop as a result of proper training. In our country, the current game of football for children aged between 8 and 10 years does not focus on the development of their creativity and coordination skills, but predominantly on the systematic tactical game following various predefined play routes. Taking into account the above, we consider it imperative to optimise the coordination skills of young players. The purpose of this research is to improve the level of coordination skills in a team of children aged 8-10 years through means specific to the game of football. During the research, two assessments were performed to determine the level of coordination skills by using the Agility T-Test (BlazePod technology) and Adams Test, which are specific for the assessment of these qualities. The research includes 42 male players aged between 8 and 10 years, as members of two football clubs, namely FC Chelsea Bucharest and FC Tineretul Bucharest. The research methods used are: documentation, experiment, recording, conversation, mathematical statistics and graphical representation. The statistically significant differences allow us to say that the improvements achieved are not random but are also due to the exercise programme intervention. As regards the T-Test specific to coordination skills, an improvement of 0.12 sec is noted for the experimental group and an improvement of only 0.05 sec for the control group.

Recreational Swimming – a Way to Improve Motor Skills

Sport has long been considered a social phenomenon that addresses all people, being a right recognised at international and national levels. Initially, adapted sport was a means of medical recovery and a means of spending free time in a pleasant way alongside peers with and without disabilities. Subsequently, adapted sport has turned into a means of comparing the motor skills of contestants having the same medical problems during competitions specially organized for these athletes. In recent years, there has been an increase in the number of people with disabilities who play sports but also in the number of non-disabled people who, in one form or another, get involved in adapted sport. In the present study, we aim to highlight the progress achieved in learning and consolidating the swimming technique by an athlete with Down syndrome (aged 25) who has been practising swimming as a recreational activity for 1 year. Based on certain items, the swimming technique was assessed for the three styles known to the athlete, and the means used in the 18 training sessions conducted over 7 months aimed at achieving the goals established at the beginning of the study. Structured, participatory and systematic observation, documentation, mathematical statistics and graphical representation are the methods used to complete the case study. At the end of the study, we could conclude that the athlete has obviously progressed, which is revealed by the assessments carried out. However, the process of learning and improvement had to be continued for the athlete to succeed in swimming correctly and be able to participate in swimming competitions without being disqualified because of technical errors.

AN IMPROVED TERM WEIGHTING AND DOCUMENT RANKING METHOD USING RANDOM WALK MODEL FOR INFORMATION RETRIEVAL

Document representation is one of the most fundamental issues in information retrieval application. The graph-based ranking algorithms represent document as a graph. Once a document is represented as graph, the similarity of that document to a query can be calculated in various ways and the calculation provides ranking to documents. This paper introduces an improved random-walk method to rank a document by considering position of a term within a document and information gain of that term within the whole document set. The experiments on various collection sets show that our approach improves the recall and precision than other proposed methods.

Estimation of Demographic Traits of the Deputies through Parliamentary Debates Using Machine Learning

One of the most impressive applications of the combined use of natural language processing (NLP), classical machine learning, and deep learning (DL) approaches is the estimation of demographic traits from the text. Author Profiling (AP) is the analysis of a text to identify the demographics or characteristics of its author. So far, most researchers in this field have focused on using social media data in the English language. This article aims to expand the predictive potential of demographic traits by focusing on a more diverse dataset and language. Knowing the background of deputies is essential for citizens, political scientists and policymakers. In this study, we present the application of NLP and machine learning (ML) approaches to Turkish parliamentary debates to estimate the demographic traits of the deputies. Seven traits were determined: gender, age, education, occupation, election region, party, and party status. As a first step, a corpus was compiled from Turkish parliamentary debates between 2012 and 2020. Document representations (feature extraction) were performed using various NLP techniques. Then, we created sub-datasets containing the extracted features from the corpus. These sub-datasets were used by different ML classification algorithms. The best classification accuracy rates were more than 31%, 27%, 35%, 41%, 29%, 59%, and 32% according to the majority baseline for gender, age, education, occupation, election region, party, and party status, respectively. The experimental results show that the demographics of deputies can be estimated effectively using NLP, classical ML, and DL approaches.

A Big Data Pipeline and Machine Learning for Uniform Semantic Representation of Data and Documents From IT Systems of the Italian Ministry of Justice

In this paper a Big Data Pipeline is presented, taking in consideration both structured and unstructured data made available by the Italian Ministry of Justice, regarding their Telematic Civil Process. Indeed, the complexity and volume of the data provided by the Ministry requires the application of Big Data analysis techniques, in concert with Machine and Deep Learning frameworks, to be correctly analysed and to obtain meaningful information that could support the Ministry itself in better managing Civil Processes. The Pipeline has two main objectives: to provide a consistent workflow of activities to be applied to the incoming data, aiming at extracting useful information for the Ministry's decision making tasks; to homogenize the incoming data, so that they can be stored in a centralized and coherent Datalake to be used as a reference for further analysis and considerations.

Identifying Adverse Drug Reaction-Related Text from Social Media: A Multi-View Active Learning Approach with Various Document Representations

Adverse drug reactions (ADRs) are a huge public health issue. Identifying text that mentions ADRs from a large volume of social media data is important. However, we need to address two challenges for high-performing ADR-related text detection: the data imbalance problem and the requirement of simultaneously using data-driven information and handcrafted information. Therefore, we propose an approach named multi-view active learning using domain-specific and data-driven document representations (MVAL4D), endeavoring to enhance the predictive capability and alleviate the requirement of labeled data. Specifically, a new view-generation mechanism is proposed to generate multiple views by simultaneously exploiting various document representations obtained using handcrafted feature engineering and by performing deep learning methods. Moreover, different from previous active learning studies in which all instances are chosen using the same selection criterion, MVAL4D adopts different criteria (i.e., confidence and informativeness) to select potentially positive instances and potentially negative instances for manual annotation. The experimental results verify the effectiveness of MVAL4D. The proposed approach can be generalized to many other text classification tasks. Moreover, it can offer a solid foundation for the ADR mention extraction task, and improve the feasibility of monitoring drug safety using social media data.

Privacy-Preserving smart contracts for Fuzzy WordNet Based Document Representation and Clustering using Regularized K-Means Method

Privacy-Preserving smart contracts for Fuzzy WordNet Based Document Representation and Clustering using Regularized K-Means Method

Document Representation and Query Expansion Models for Blog Recommendation

We explore several different document representation models and two query expansion models for the task of recommending blogs to a user in response to a query. Blog relevance ranking differs from traditional document ranking in ad-hocinformation retrieval in several ways: (1) the unit of output (the blog) is composed of a collection of documents (the blog posts) rather than a single document, (2) the query represents an ongoing and typically multifaceted interest in the topic rather than a passing ad-hoc information need and (3) due to the propensity of spam, splogs, and tangential comments, the blogosphere is particularly challenging to use as a source for high-quality query expansion terms. We address these differences at the document representation level, by comparing retrieval models that view either the blog or its constituent posts as the atomic units of retrieval, and at the query expansion level, by making novel use of the links and anchor text in Wikipedia1 to expand a user's initial query. We develop two complementary models of blog retrieval that perform at comparable levels of precision and recall. We also show consistent and significant improvement across all models using our Wikipedia expansion strategy.

Mining on the frontier: Archaeological excavation of the historical component at Canteen Kopje, Northern Cape Province, South Africa

Canteen Kopje was a prominent location within the spread of nineteenth century mining camps at Klipdrift along the Vaal River. This article presents the results of excavation of the historical component of Canteen Kopje that produced an artifact assemblage in which European manufactured ceramics and glass are associated with objects of local manufacture, which attests to the cultural interaction that took place in the context of early mining. These finds of material culture are associated with a faunal assemblage of wild animals primarily the remains of quagga, a species that was soon to go in extinct. Based on historical documentation and pictorial representation it is argued that the earliest Vaal River diamond diggings here represent a stage of extractive economies in southern Africa when the rigid control of space and people that emerged subsequently was not yet fully developed.

A Graph Database Representation of Portuguese Criminal-Related Documents

Organizations have been challenged by the need to process an increasing amount of data, both structured and unstructured, retrieved from heterogeneous sources. Criminal investigation police are among these organizations, as they have to manually process a vast number of criminal reports, news articles related to crimes, occurrence and evidence reports, and other unstructured documents. Automatic extraction and representation of data and knowledge in such documents is an essential task to reduce the manual analysis burden and to automate the discovering of names and entities relationships that may exist in a case. This paper presents SEMCrime, a framework used to extract and classify named-entities and relations in Portuguese criminal reports and documents, and represent the data retrieved into a graph database. A 5WH1 (Who, What, Why, Where, When, and How) information extraction method was applied, and a graph database representation was used to store and visualize the relations extracted from the documents. Promising results were obtained with a prototype developed to evaluate the framework, namely a name-entity recognition with an F-Measure of 0.73, and a 5W1H information extraction performance with an F-Measure of 0.65.

Adaptive and hybrid context-aware fine-grained word sense disambiguation in topic modeling based document representation

We propose a hybrid context based topic model with an adaptive context window length for word sense disambiguation in document representation. Document representation is an essential part of various document based tasks, and word sense disambiguation is to capture the distinctions of word senses in the representation. Traditional methods mainly rely on knowledge libraries for data enrichment; however, semantics division for a word may vary in different domain-specific datasets. We aim to discover finer-grained word semantic differences, such as different entities or standpoints, and handle the disambiguation problem without data enrichment. There are two challenges for this disambiguation task: (1) dividing various senses for each polysemous word, and (2) preserving the differences between synonyms. Most of the existing models are either based on separate context clusters or integrating an auxiliary module to specify word senses. They can hardly achieve both (1) and (2) since different senses of a word are assumed to be independent and their intrinsic relationships are ignored. To solve this problem, we introduce the “Bag-of-Senses” (BoS) assumption: a document is a multiset of word senses, and the senses are generated instead of the words. The word senses are estimated by both the context in which it occurs and the contexts of its other occurrences. Besides, to handle the different scopes of the sense related context to each word occurrence, we introduce a variable to adjust the context window length adaptively. Our experiments on three standard datasets show that our proposal outperforms other state-of-the-art methods in terms of word sense estimation, topic modeling, and document classification.

Vector representation based on a supervised codebook for Nepali documents classification.

Document representation with outlier tokens exacerbates the classification performance due to the uncertain orientation of such tokens. Most existing document representation methods in different languages including Nepali mostly ignore the strategies to filter them out from documents before learning their representations. In this article, we propose a novel document representation method based on a supervised codebook to represent the Nepali documents, where our codebook contains only semantic tokens without outliers. Our codebook is domain-specific as it is based on tokens in a given corpus that have higher similarities with the class labels in the corpus. Our method adopts a simple yet prominent representation method for each word, called probability-based word embedding. To show the efficacy of our method, we evaluate its performance in the document classification task using Support Vector Machine and validate against widely used document representation methods such as Bag of Words, Latent Dirichlet allocation, Long Short-Term Memory, Word2Vec, Bidirectional Encoder Representations from Transformers and so on, using four Nepali text datasets (we denote them shortly as A1, A2, A3 and A4). The experimental results show that our method produces state-of-the-art classification performance (77.46% accuracy on A1, 67.53% accuracy on A2, 80.54% accuracy on A3 and 89.58% accuracy on A4) compared to the widely used existing document representation methods. It yields the best classification accuracy on three datasets (A1, A2 and A3) and a comparable accuracy on the fourth dataset (A4). Furthermore, we introduce the largest Nepali document dataset (A4), called NepaliLinguistic dataset, to the linguistic community.

Interactive virtual representation of the disappeared convent of El Carmen (Logroño) generated from a paper craft model

<p class="VARAbstract">The concept of mock-up, which share with the drawing the expressive synthesis for the analysis of the architectural form and evolution, is an invaluable asset for the preservation of the heritage. To mark the 175<sup>th</sup> anniversary of the Práxedes Mateo Sagasta’s secondary school of Logroño (Spain), an exhibition about the history of the institution was organized at the premises of La Rioja Library. The current school building became operational in 1900 and was built on the former site of a Carmelite convent. In fact, the convent rooms were the first location for the school, after the expropriations of religious communities during the mid-19th century. For the benefit of the aforementioned exhibition, it was considered interesting to generate a three-dimensional (3D) virtual reconstruction of the convent buildings to show how it would have looked. However, the lack of sufficient contemporary graphic information was a challenge, so it was decided that an efficient solution would be to generate the virtual reconstruction from a paper craft model, which had been created by the librarian of the secondary school as a result of his research on this matter. This text describes the 3D modelling of that cut-out element by means of monoscopic photogrammetry (perspective drawing) and the use of non-realistic rendering based on the appearance of the paper mock-up (and not trying to recreate the real image of the buildings) so as to provide a suggestive view of the convent and create interactive exhibition items. Moreover, the text deals with the long-term preservation and the improvement of the re-use of the 3D models. The former by resorting to institutional repositories —from which users can download the full detailed versions— and the latter employing versions able to be visualized in 3D warehouses (such as Sketchfab) and augmented reality (AR) applications.</p><p><strong>Highlights:</strong></p><ul><li><p>A significant part of the architectural heritage is represented by mock-ups of different materials (paper, wood, metal and so on).</p></li><li><p>Virtual modelling and augmented reality (AR) can be appropriate tools for materializing, recovering and disseminating scale models to the public.</p></li><li><p>Techniques for geometric documentation and visual representation need to be adapted in order to tackle the peculiar features of these elements (size, materials, fragility, etc.).</p></li></ul>

A Hybrid Model for Documents Representation

Text representation is a critical issue for exploring the insights behind the text. Many models have been developed to represent the text in defined forms such as numeric vectors where it would be easy to calculate the similarity between the documents using the well-known distance measures. In this paper, we aim to build a model to represent text semantically either in one document or multiple documents using a combination of hierarchical Latent Dirichlet Allocation (hLDA), Word2vec, and Isolation Forest models. The proposed model aims to learn a vector for each document using the relationship between its words’ vectors and the hierarchy of topics generated using the hierarchical Latent Dirichlet Allocation model. Then, the isolation forest model is used to represent multiple documents in one representation as one profile to facilitate finding similar documents to the profile. The proposed text representation model outperforms the traditional text representation models when applied to represent scientific papers before performing content-based scientific papers recommendation for researchers.

INFLUENCE OF THE INSANITY WORKOUT ON THE EXERCISE CAPACITY OF MILITARY STUDENTS

The research purpose is to improve the exercise capacity of military students (VAM, VO2 max and average heart rate) by applying the Insanity Workout in the physical education lesson. The following scientific methods were used: documentation, observation, statistical processing and graphical representation of the results. Subject assessment was based on the Yo-Yo Intermittent Recovery Test Level 1 (YYIR1), and Polar H10 Heart Rate Sensor was used. The research included the following stages: stage 1 - conducting an initial assessment; stage 2 - applying the Insanity Workout; stage 3 - performing a final assessment of the studied subjects to assess their response (evolution/involution/stagnation); stage 4 - interpretation of results and conclusion. Based on the paired t-test, the following results are obtained: the increase of the average maximum aerobic speed achieved at the end of the training period is statistically significant, p = 0.005 < 0.05 for t = 3.003 and df = 29. The effect size index indicates a medium-to-high difference between the two means; the increase of the average maximum aerobic capacity achieved at the end of the training period is statistically significant, p < 0.001 < 0.05 for t = 6.536 and df = 29. The effect size index indicates a significant difference between the two means; the decrease of the average heart rate at the end of the training period is statistically significant, p < 0.001 < 0.05 for t = 5.012 and df = 29. The effect size index indicates a high difference between the two means.

STUDY ON LEARNING THE BOCCE-SPECIFIC TECHNIQUE

The truth that the physical exercises practiced either independently or systematically has beneficial effects on the whole body has long been demonstrated and supported by empirical or scientific studies. Subsequently, this assertion was also proved in the case of people with intellectual disabilities. For this category of population, the level of learning various motor skills is different from one individual to another. There are periods when the learning is very fast, the pace being close to that of valid individuals. However, these periods alternate with periods of stagnation or even regression. In the learning phase, successful attempts are rare and technically incorrect. Gradually, people with Down syndrome manage to learn the technique specific to a sport discipline and commonly use it according to their needs and the specific circumstances. In this context, the present study aimed to identify how athletes with Down syndrome have learned and enhanced the technique specific to the game of bocce, especially the technique of releasing the ball. For this, we tested 16 athletes with Down syndrome (10 children and 6 adults; 8 girls and 8 boys) who were members of the target group included in the Erasmus+ Sport 590526 project. The methods used to conduct the study were: documentation, experiment, statistics and graphical representation. The test applied to the research subjects, namely the divisioning bocce test, resulted in the recording of higher performances for 10 athletes, while the performances of the other athletes were not statistically significant.

Self-Supervised Document Representation Method

Self-Supervised Document Representation Method

Combining Distributed Word Representation and Document Distance for Short Text Document Clustering

This paper presents a method for clustering short text documents, such as news headlines, social media statuses, or instant messages. Due to the characteristics of these documents, which are usually short and sparse, an appropriate technique is required to discover hidden knowledge. The objective of this paper is to identify the combination of document representation, document distance, and document clustering that yields the best clustering quality. Document representations are expanded by external knowledge sources represented by a Distributed Representation. To cluster documents, a K-means partitioning-based clustering technique is applied, where the similarities of documents are measured by word mover’s distance. To validate the effectiveness of the proposed method, experiments were conducted to compare the clustering quality against several leading methods. The proposed method produced clusters of documents that resulted in higher precision, recall, F1- score, and adjusted Rand index for both real-world and standard data sets. Furthermore, manual inspection of the clustering results was conducted to observe the efficacy of the proposed method. The topics of each document cluster are undoubtedly reflected by members in the cluster.