Search results

Hundreds of thousands of deaths each year in the world are caused by breast cancer (BC). An early-stage diagnosis of this disease can positively reduce the morbidity and mortality rate by helping to select the most appropriate treatment options, especially by using histological BC images for the diagnosis.

The present study proposes and evaluates a novel approach which consists of 24 deep hybrid heterogenous ensembles that combine the strength of seven deep learning techniques (DenseNet 201, Inception V3, VGG16, VGG19, Inception-ResNet-V3, MobileNet V2 and ResNet 50) for feature extraction and four well-known classifiers (multi-layer perceptron, support vector machines, K-nearest neighbors and decision tree) by means of hard and weighted voting combination methods for histological classification of BC medical image. Furthermore, the best deep hybrid heterogenous ensembles were compared to the deep stacked ensembles to determine the best strategy to design the deep ensemble methods. The empirical evaluations used four classification performance criteria (accuracy, sensitivity, precision and F1-score), fivefold cross-validation, Scott–Knott (SK) statistical test and Borda count voting method. All empirical evaluations were assessed using four performance measures, including accuracy, precision, recall and F1-score, and were over the histological BreakHis public dataset with four magnification factors (40×, 100×, 200× and 400×). SK statistical test and Borda count were also used to cluster the designed techniques and rank the techniques belonging to the best SK cluster, respectively.

Results showed that the deep hybrid heterogenous ensembles outperformed both their singles and the deep stacked ensembles and reached the accuracy values of 96.3, 95.6, 96.3 and 94 per cent across the four magnification factors 40×, 100×, 200× and 400×, respectively.

The proposed deep hybrid heterogenous ensembles can be applied for the BC diagnosis to assist pathologists in reducing the missed diagnoses and proposing adequate treatments for the patients.

This study aims to report on an automatic deidentification system for labeling and obfuscating personally identifiable information (PII) in student-generated text.

The authors evaluate the performance of their deidentification system on two data sets of student-generated text. Each data set was human-annotated for PII. The authors evaluate using two approaches: per-token PII classification accuracy and a simulated reidentification attack design. In the reidentification attack, two reviewers attempted to recover student identities from the data after PII was obfuscated by the authors’ system. In both cases, results are reported in terms of recall and precision.

The authors’ deidentification system recalled 84% of student name tokens in their first data set (96% of full names). On the second data set, it achieved a recall of 74% for student name tokens (91% of full names) and 75% for all direct identifiers. After the second data set was obfuscated by the authors’ system, two reviewers attempted to recover the identities of students from the obfuscated data. They performed below chance, indicating that the obfuscated data presents a low identity disclosure risk.

The two data sets used in this study are not representative of all forms of student-generated text, so further work is needed to evaluate performance on more data.

This paper presents an open-source and automatic deidentification system appropriate for student-generated text with technical explanations and evaluations of performance.

Previous study on text deidentification has shown success in the medical domain. This paper develops on these approaches and applies them to text in the educational domain.

The purpose of this paper is to present a preliminary work on extracting band gap information of materials from academic papers. With increasing demand for renewable energy, band gap information will help material scientists design and implement novel photovoltaic (PV) cells.

The authors collected 1.44 million titles and abstracts of scholarly articles related to materials science, and then filtered the collection to 11,939 articles that potentially contain relevant information about materials and their band gap values. ChemDataExtractor was extended to extract information about PV materials and their band gap information. Evaluation was performed on randomly sampled information records of 415 papers.

The findings of this study show that the current system is able to correctly extract information for 51.32% articles, with partially correct extraction for 36.62% articles and incorrect for 12.04%. The authors have also identified the errors belonging to three main categories pertaining to chemical entity identification, band gap information and interdependency resolution. Future work will focus on addressing these errors to improve the performance of the system.

The authors did not find any literature to date on band gap information extraction from academic text using automated methods. This work is unique and original. Band gap information is of importance to materials scientists in applications such as solar cells, light emitting diodes and laser diodes.

This chapter explores the potential for a rigid format of event structure that has been successful for decades to be curated by, hosted by, attended by and with speakers who are all children or young people. The chapter considers the opportunities for an event and the changes in format that need to be made to stay true to the essence of the TED format yet incorporating the creativity and brilliance of children and young people to ensure they gain the most from an event of this format.

The purpose of this study is to solve the problems caused by the growing volumes of pre-annotated literature and variety-oriented annotations, including teamwork, quality control and time effort.

An annotation collaboration workbench is developed, which is named as Bureau for Rapid Annotation Tool (Brat). Main functionalities include an enhanced semantic constraint system, Vim-like shortcut keys, an annotation filter and a graph-visualizing annotation browser. With these functionalities, the annotators are encouraged to question their initial mindset, inspect conflicts and gain agreement from their peers.

The collaborative patterns can indeed be leveraged to structure properly every annotator’s behaviors. The Brat workbench can actually be seen as an experienced-based annotation tool by harnessing collective intelligence. Compared to previous counterparts, about one-third of time can be saved on Xinhuanet military news and patent corpora with the workbench.

The various annotations are very popular in real-world annotation tasks with multiple annotators. Though, it is still under-discussed on variety-oriented annotations. The findings of this study provide the practitioners valuable insight into how to govern annotation projects. In addition, the Brat workbench takes the first step for future research on annotating large-scale text resources.

Building on past studies of library privacy policies, this review looks at how privacy information is shared at universities and colleges in the state of Florida. Beyond the question of whether a library-specific privacy policy exists, this review evaluates what is covered in the policies – whether topics such as how student data is stored, retained, de-identified and disposed of are broached in the statements, and whether specific data sets covering instruction, reference and surveillance are mentioned. The purpose of this study is to open the door to directed exploration into student awareness of privacy policies and spark conversation about positionality of libraries regarding privacy.

This review was done using a cross-sectional study design through observation of public-facing library privacy policies of higher education institutions in Florida.

Findings include that the majority of Florida academic libraries do not have a public-facing privacy policy. Only 15 out of the 70 schools reviewed had one. A large portion of those came from doctoral universities with associate’s colleges having none, and baccalaureate/associate’s colleges having only two. The policies that were in place tended to be institution-centered rather than patron-centered. Most categories of listed data collected were in the area of collections, website or computer usage.

The value of this review is that it adds to the literature studying privacy policies in academic libraries. Going forward, this research could address statewide practice in privacy policies as well as helping to lay pathways for working with students and other library patrons to gauge their interests and concerns about privacy.

With the development of data mining technology, diverse and broader domain knowledge can be extracted automatically. However, the research on applying knowledge mapping and data visualization techniques to genealogical data is limited. This paper aims to fill this research gap by providing a systematic framework and process guidance for practitioners seeking to uncover hidden knowledge from genealogy.

Based on a literature review of genealogy's current knowledge reasoning research, the authors constructed an integrated framework for knowledge inference and visualization application using a knowledge graph. Additionally, the authors applied this framework in a case study using “Manchu Clan Genealogy” as the data source.

The case study shows that the proposed framework can effectively decompose and reconstruct genealogy. It demonstrates the reasoning, discovery, and web visualization application process of implicit information in genealogy. It enhances the effective utilization of Manchu genealogy resources by highlighting the intricate relationships among people, places, and time entities.

This study proposed a framework for genealogy knowledge reasoning and visual analysis utilizing a knowledge graph, including five dimensions: the target layer, the resource layer, the data layer, the inference layer, and the application layer. It helps to gather the scattered genealogy information and establish a data network with semantic correlations while establishing reasoning rules to enable inference discovery and visualization of hidden relationships.

The aim of this paper is to systematically review the literature published in recognized journals focused on recognition-based heuristics and their effect on investment management activities and to ascertain some substantial gaps related to them.

For doing research synthesis, systematic literature review approach was applied considering research studies published within the time period, i.e. 1980–2020. This study attempted to accomplish a critical review of 59 studies out of 118 studies identified, which were published in reputable journals to synthesize the existing literature in the behavioural finance domain-related explicitly to recognition-based heuristics and their effect on investment management activities.

The survey and analysis suggest investors consistently rely on the recognition-based heuristic-driven biases when trading stocks, resulting in irrational decisions, and an investment strategy constructed by implementing the recognition-based heuristics, would not result in better returns to investors on a consistent basis. Institutional investors are less likely to be affected by these name-based behavioural biases in comparison to individual investors. However, under the context of ecological rationality, recognition-based heuristics work better and sometimes dominate the classical methods. The research scholars from the behavioural finance community have highlighted that recognition-based heuristics and their impact on investment management activities are high profile areas, needed to be explored further in the field of behavioural finance. The study of recognition-based heuristic-driven biases has been found to be insufficient in the context of emerging economies like Pakistan.

The skilful understanding and knowledge of the recognition-based heuristic-driven biases will help the investors, financial institutions and policy-makers to overcome the adverse effect of these behavioural biases in the stock market. This article provides a detailed explanation of recognition-based heuristic-driven biases and their influence on investment management activities which could be very useful for finance practitioners’ such as investor who plays at the stock exchange, a portfolio manager, a financial strategist/advisor in an investment firm, a financial planner, an investment banker, a trader/ broker at the stock exchange or a financial analyst. But most importantly, the term also includes all those persons who manage corporate entities and are responsible for making its financial management strategies.

Currently, no recent study exists, which reviews and evaluates the empirical research on recognition-based heuristic-driven biases displayed by investors. The current study is original in discussing the role of recognition-based heuristic-driven biases in investment management activities by means of research synthesis. This paper is useful to researchers, academicians, and those working in the area of behavioural finance in understanding the role that recognition-based heuristics plays in investment management activities.

The purpose of this paper is to propose a methodology for the enrichment and tailoring of a knowledge organization system (KOS), in order to support the information extraction (IE) task for the analysis of documents in the tourism domain. In particular, the KOS is used to develop a named entity recognition (NER) system.

A method to improve and customize an available thesaurus by leveraging documents related to the tourism in Italy is firstly presented. Then, the obtained thesaurus is used to create an annotated NER corpus, exploiting both distant supervision, deep learning and a light human supervision.

The study shows that a customized KOS can effectively support IE tasks when applied to documents belonging to the same domains and types used for its construction. Moreover, it is very useful to support and ease the annotation task using the proposed methodology, allowing to annotate a corpus with a fraction of the effort required for a manual annotation.

The paper explores an alternative use of a KOS, proposing an innovative NER corpus annotation methodology. Moreover, the KOS and the annotated NER data set will be made publicly available.

Explore the factors making emergency procurement more prone to corruption by advancing explanations for when rules and transparency are relaxed allowing corrupt practices to emerge. Describe institutional factors, such as corruption syndrome (Johnston, 2005, 2015) and legal system, and their impact on procurement rules changes.

A qualitative event study using publicly available data offer a timeline and explanation of government procurement control mechanisms and transparency roles in emergencies by comparing two countries. Argentina and Canada had very similar and advanced food procurement systems prior to COVID-19, but they took different stances when the pandemic broke out.

Legal systems and corruption syndrome are linked, where Civil Law is related to Elite Cartels (Argentina) and Common Law with Influence Markets (Canada). The study contributes to understand the role of transparency to minimize the opportunity for direct purchases (electronic trails of decisions, justifications and approvals). Judicial system's actions favor corrupt practices and are aligned with elites despite the civil society outcry.

Research on corrupt practices has limited access to primary data due to fear of reprisals. Informal conversations revealing glimpses of corruption were used to identify publicly available documents. Numbers play a role in emergencies and performativity theory literature is enriched by providing an example of different interpretation of information when frameworks differ between civil society and courts.

A comparative analysis that evidences the role of pre-existing institutional and social conditions shows when emergency situations will be used as an excuse to relax procurement control and transparency mechanisms which in turn facilitate corrupt practices.