The Influence of Surface Texture of Elements Made of PA6-Based Composites on Anti-Graffiti Effect of Paint Coating.

The aim of the work was to investigate the influence of the surface texture of composite based on PA6, intended for wet painting, on the stability of the colour and gloss parameters. The stability of the paint coating was required to be maintained despite exposure to mechanical stress resulting from attempts to manually remove graffiti stains. The study examined the influence of surface texture on the effectiveness of cleaning. In the case of painted surfaces from which graffiti stains were effectively removed, the roughness, colour parameters and gloss of the paint coating were measured. During the research, it was found that roughness after painting decreased to the value of Ra < 2.00 µm meets aesthetic expectations and at the same time ensures the effective removal of graffiti stains. For this surface, there were no negative effects of the mechanical impact on the textures or quality parameters of the coating as a result of manual graffiti removal. As a result of the conducted research, the recommended maximum values of roughness and textures of the surfaces to be painted were determined in order to ensure a sufficiently low amount of work necessary to effectively remove traces of graffiti.

Deep Learning for Combating Misinformation in Multicategorical Text Contents.

Currently, one can observe the evolution of social media networks. In particular, humans are faced with the fact that, often, the opinion of an expert is as important and significant as the opinion of a non-expert. It is possible to observe changes and processes in traditional media that reduce the role of a conventional 'editorial office', placing gradual emphasis on the remote work of journalists and forcing increasingly frequent use of online sources rather than actual reporting work. As a result, social media has become an element of state security, as disinformation and fake news produced by malicious actors can manipulate readers, creating unnecessary debate on topics organically irrelevant to society. This causes a cascading effect, fear of citizens, and eventually threats to the state's security. Advanced data sensors and deep machine learning methods have great potential to enable the creation of effective tools for combating the fake news problem. However, these solutions often need better model generalization in the real world due to data deficits. In this paper, we propose an innovative solution involving a committee of classifiers in order to tackle the fake news detection challenge. In that regard, we introduce a diverse set of base models, each independently trained on sub-corpora with unique characteristics. In particular, we use multi-label text category classification, which helps formulate an ensemble. The experiments were conducted on six different benchmark datasets. The results are promising and open the field for further research.

Sensors and Pattern Recognition Methods for Security and Industrial Applications.

Contemporary cyberthreats continue to evolve, powering the neverending development arms race [...].

Has the pandemic made us more digitally literate?: Innovative association rule mining study of the relationships between shifts in digital skills and cybersecurity awareness occurring whilst working remotely during the COVID-19 pandemic.

Digital literacy has been included in the set of the eight key competences, which are necessary to enjoy life to the full in the twenty-first century. According to the previous studies, women tend to possess lower digital competence than men; the older the person, the lower the level of digital literacy. To date, Polish citizens in general have worse skills than the European average. This may lead to people being socially excluded and vulnerable to cybersecurity threats, especially in the times of the COVID-19 pandemic, which requires them to work, study and shop using the Internet. The study concerned Polish women who work at universities, as scientists and teachers. Their perceived level of their digital literacy has been studied in the broad campaign, along with their awareness of the cybersecurity matters. Then, the collected results were processed with an association rules mining algorithm, uncovering the factors related to the shifts in them.

Opening up a discussion on the native speaker myth in science.

The commentary touches upon the topic which is relevant to hundreds of thousands of researchers in the world. When trying to publish in English, they are often advised to ask the native speakers of the language for their opinion. However, as English has become the international, cross-border language of science, it may have ceased to be the property of the native speaker researchers, who constitute a small minority in the community. In addition, when English is used as a lingua franca, it is the message which counts, not the particular style or spelling. The commentary finishes with an appeal not to hinder the development of science by slowing down the editing process, and thus not to close the door for diversity and new perspectives.

New explainability method for BERT-based model in fake news detection.

The ubiquity of social media and their deep integration in the contemporary society has granted new ways to interact, exchange information, form groups, or earn money-all on a scale never seen before. Those possibilities paired with the widespread popularity contribute to the level of impact that social media display. Unfortunately, the benefits brought by them come at a cost. Social Media can be employed by various entities to spread disinformation-so called 'Fake News', either to make a profit or influence the behaviour of the society. To reduce the impact and spread of Fake News, a diverse array of countermeasures were devised. These include linguistic-based approaches, which often utilise Natural Language Processing (NLP) and Deep Learning (DL). However, as the latest advancements in the Artificial Intelligence (AI) domain show, the model's high performance is no longer enough. The explainability of the system's decision is equally crucial in real-life scenarios. Therefore, the objective of this paper is to present a novel explainability approach in BERT-based fake news detectors. This approach does not require extensive changes to the system and can be attached as an extension for operating detectors. For this purposes, two Explainable Artificial Intelligence (xAI) techniques, Local Interpretable Model-Agnostic Explanations (LIME) and Anchors, will be used and evaluated on fake news data, i.e., short pieces of text forming tweets or headlines. This focus of this paper is on the explainability approach for fake news detectors, as the detectors themselves were part of previous works of the authors.

How to Effectively Collect and Process Network Data for Intrusion Detection?

The number of security breaches in the cyberspace is on the rise. This threat is met with intensive work in the intrusion detection research community. To keep the defensive mechanisms up to date and relevant, realistic network traffic datasets are needed. The use of flow-based data for machine-learning-based network intrusion detection is a promising direction for intrusion detection systems. However, many contemporary benchmark datasets do not contain features that are usable in the wild. The main contribution of this work is to cover the research gap related to identifying and investigating valuable features in the NetFlow schema that allow for effective, machine-learning-based network intrusion detection in the real world. To achieve this goal, several feature selection techniques have been applied on five flow-based network intrusion detection datasets, establishing an informative flow-based feature set. The authors' experience with the deployment of this kind of system shows that to close the research-to-market gap, and to perform actual real-world application of machine-learning-based intrusion detection, a set of labeled data from the end-user has to be collected. This research aims at establishing the appropriate, minimal amount of data that is sufficient to effectively train machine learning algorithms in intrusion detection. The results show that a set of 10 features and a small amount of data is enough for the final model to perform very well.

A $10 million question and other cybersecurity-related ethical dilemmas amid the COVID-19 pandemic.

Cybercrime and cybersecurity are like two sides of the same coin: They are opposites but cannot exist without each other. Their mutual relation generates a myriad of ethical issues, ranging from minor to vital. The rapid development of technology will surely involve even more ethical concerns, like the infamous example of a fitness tracking company allegedly paying $10 million worth of ransom. Every cybersecurity solution, tool, or practice has to be ethical by design if it is to protect people and their rights. To identify the ethical issues that cybersecurity/cybercrime might bring about in the future, we conducted the first broad and comprehensive horizon-scanning study since the COVID-19 pandemic arose. As we began this project, nobody had the slightest idea that the coming months would bring the COVID-19 pandemic, and that the reality we had known was about to change dramatically. As it soon became apparent, the deadly coronavirus brought completely new cybersecurity/cybercrime ethical dilemmas to light, and some of the ones known before were transformed or shifted. This article presents the results of our horizon-scanning study concerning the ethical dilemmas that emerged amid the COVID-19 pandemic.

Preprocessing Pipelines including Block-Matching Convolutional Neural Network for Image Denoising to Robustify Deep Reidentification against Evasion Attacks.

Artificial neural networks have become the go-to solution for computer vision tasks, including problems of the security domain. One such example comes in the form of reidentification, where deep learning can be part of the surveillance pipeline. The use case necessitates considering an adversarial setting-and neural networks have been shown to be vulnerable to a range of attacks. In this paper, the preprocessing defences against adversarial attacks are evaluated, including block-matching convolutional neural network for image denoising used as an adversarial defence. The benefit of using preprocessing defences comes from the fact that it does not require the effort of retraining the classifier, which, in computer vision problems, is a computationally heavy task. The defences are tested in a real-life-like scenario of using a pre-trained, widely available neural network architecture adapted to a specific task with the use of transfer learning. Multiple preprocessing pipelines are tested and the results are promising.

A Systematic Review of Recommender Systems and Their Applications in Cybersecurity.

This paper discusses the valuable role recommender systems may play in cybersecurity. First, a comprehensive presentation of recommender system types is presented, as well as their advantages and disadvantages, possible applications and security concerns. Then, the paper collects and presents the state of the art concerning the use of recommender systems in cybersecurity; both the existing solutions and future ideas are presented. The contribution of this paper is two-fold: to date, to the best of our knowledge, there has been no work collecting the applications of recommenders for cybersecurity. Moreover, this paper attempts to complete a comprehensive survey of recommender types, after noticing that other works usually mention two-three types at once and neglect the others.

The Proposition and Evaluation of the RoEduNet-SIMARGL2021 Network Intrusion Detection Dataset.

Cybersecurity is an arms race, with both the security and the adversaries attempting to outsmart one another, coming up with new attacks, new ways to defend against those attacks, and again with new ways to circumvent those defences. This situation creates a constant need for novel, realistic cybersecurity datasets. This paper introduces the effects of using machine-learning-based intrusion detection methods in network traffic coming from a real-life architecture. The main contribution of this work is a dataset coming from a real-world, academic network. Real-life traffic was collected and, after performing a series of attacks, a dataset was assembled. The dataset contains 44 network features and an unbalanced distribution of classes. In this work, the capability of the dataset for formulating machine-learning-based models was experimentally evaluated. To investigate the stability of the obtained models, cross-validation was performed, and an array of detection metrics were reported. The gathered dataset is part of an effort to bring security against novel cyberthreats and was completed in the SIMARGL project.

A Deep Learning Ensemble for Network Anomaly and Cyber-Attack Detection.

Currently, expert systems and applied machine learning algorithms are widely used to automate network intrusion detection. In critical infrastructure applications of communication technologies, the interaction among various industrial control systems and the Internet environment intrinsic to the IoT technology makes them susceptible to cyber-attacks. Given the existence of the enormous network traffic in critical Cyber-Physical Systems (CPSs), traditional methods of machine learning implemented in network anomaly detection are inefficient. Therefore, recently developed machine learning techniques, with the emphasis on deep learning, are finding their successful implementations in the detection and classification of anomalies at both the network and host levels. This paper presents an ensemble method that leverages deep models such as the Deep Neural Network (DNN) and Long Short-Term Memory (LSTM) and a meta-classifier (i.e., logistic regression) following the principle of stacked generalization. To enhance the capabilities of the proposed approach, the method utilizes a two-step process for the apprehension of network anomalies. In the first stage, data pre-processing, a Deep Sparse AutoEncoder (DSAE) is employed for the feature engineering problem. In the second phase, a stacking ensemble learning approach is utilized for classification. The efficiency of the method disclosed in this work is tested on heterogeneous datasets, including data gathered in the IoT environment, namely IoT-23, LITNET-2020, and NetML-2020. The results of the evaluation of the proposed approach are discussed. Statistical significance is tested and compared to the state-of-the-art approaches in network anomaly detection.

Innovative machine learning approach and evaluation campaign for predicting the subjective feeling of work-life balance among employees.

At present, many researchers see hope that artificial intelligence, machine learning in particular, will improve several aspects of the everyday life for individuals, cities and whole nations alike. For example, it has been speculated that the so-called machine learning could soon relieve employees of part of the duties, which may improve processes or help to find the most effective ways of performing tasks. Consequently, in the long run, it would help to enhance employees' work-life balance. Thus, workers' overall quality of life would improve, too. However, what would happen if machine learning as such were employed to try and find the ways of achieving work-life balance? This is why the authors of the paper decided to utilize a machine learning tool to search for the factors that influence the subjective feeling of one's work-life balance. The possible results could help to predict and prevent the occurrence of work-life imbalance in the future. In order to do so, the data provided by an exceptionally sizeable group of 800 employees was utilised; it was one of the largest sample groups in similar studies in Poland so far. Additionally, this was one of the first studies where so many employees had been analysed using an artificial neural network. In order to enable replicability of the study, the specific setup of the study and the description of the dataset are provided. Having analysed the data and having conducted several experiments, the correlations between some factors and work-life balance have indeed been identified: it has been found that the most significant was the relation between the feeling of balance and the actual working hours; shifting it resulted in the tool predicting the switch from balance to imbalance, and vice versa. Other factors that proved significant for the predicted WLB are the amount of free time a week the employee has for themselves, working at weekends only, being self-employed and the subjective assessment of one's financial status. In the study the dataset gets balanced, the most important features are selected with the selectKbest algorithm, an artificial neural network of 2 hidden layers with 50 and 25 neurons, ReLU and ADAM is constructed and trained on 90% of the dataset. In tests, it predicts WLB based on the prepared dataset and selected features with 81% accuracy.

Health-related quality of life with adjuvant docetaxel- and trastuzumab-based regimens in patients with node-positive and high-risk node-negative, HER2-positive early breast cancer: results from the BCIRG 006 Study.

BACKGROUND: This study aims to describe and compare health-related quality of life (HRQL) in patients with node-positive and high-risk node-negative HER2-positive early breast cancer receiving adjuvant docetaxel and trastuzumab-based or docetaxel-based regimens alone. METHODS: Eligible patients (n = 3,222) were randomly assigned to either four cycles of adjuvant doxorubicin and cyclophosphamide followed by four cycles of docetaxel (AC→T) or one of two trastuzumab-containing regimens: adjuvant doxorubicin and cyclophosphamide followed by docetaxel plus trastuzumab administered for 1 year (AC→TH) or six cycles of docetaxel plus carboplatin combined with trastuzumab administered for 1 year (TCH). The European Organization for Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire C30 and BR-23 were administered at baseline, the start of cycle 4 (mid), and the end of chemotherapy (EOC), as well as at 6, 12, and 24 months after chemotherapy. RESULTS: Compliance rates for the EORTC questionnaires were acceptable at 72%-93% of eligible patients out to the 12-month assessment. Systemic side effect (SE) change scores were significantly improved for TCH-treated patients compared with AC→TH and AC→T at EOC, suggesting improved tolerability. Physical functioning (PF) was only slightly worse at midpoint for those receiving TCH, compared with patients who were just starting on taxane in an AC→TH regimen, but was otherwise similar between arms. All treatment arms recovered from the deterioration in SE, PF, and Global Health Scale scores by 1 year and median future perspective change scores continued to improve throughout treatment and follow-up. CONCLUSION: HRQL outcomes for adjuvant docetaxel and trastuzumab-based regimens are favorable and support TCH as a more tolerable treatment option.

Adjuvant trastuzumab in HER2-positive breast cancer.

BACKGROUND: Trastuzumab improves survival in the adjuvant treatment of HER-positive breast cancer, although combined therapy with anthracycline-based regimens has been associated with cardiac toxicity. We wanted to evaluate the efficacy and safety of a new nonanthracycline regimen with trastuzumab. METHODS: We randomly assigned 3222 women with HER2-positive early-stage breast cancer to receive doxorubicin and cyclophosphamide followed by docetaxel every 3 weeks (AC-T), the same regimen plus 52 weeks of trastuzumab (AC-T plus trastuzumab), or docetaxel and carboplatin plus 52 weeks of trastuzumab (TCH). The primary study end point was disease-free survival. Secondary end points were overall survival and safety. RESULTS: At a median follow-up of 65 months, 656 events triggered this protocol-specified analysis. The estimated disease-free survival rates at 5 years were 75% among patients receiving AC-T, 84% among those receiving AC-T plus trastuzumab, and 81% among those receiving TCH. Estimated rates of overall survival were 87%, 92%, and 91%, respectively. No significant differences in efficacy (disease-free or overall survival) were found between the two trastuzumab regimens, whereas both were superior to AC-T. The rates of congestive heart failure and cardiac dysfunction were significantly higher in the group receiving AC-T plus trastuzumab than in the TCH group (P<0.001). Eight cases of acute leukemia were reported: seven in the groups receiving the anthracycline-based regimens and one in the TCH group subsequent to receiving an anthracycline outside the study. CONCLUSIONS: The addition of 1 year of adjuvant trastuzumab significantly improved disease-free and overall survival among women with HER2-positive breast cancer. The risk-benefit ratio favored the nonanthracycline TCH regimen over AC-T plus trastuzumab, given its similar efficacy, fewer acute toxic effects, and lower risks of cardiotoxicity and leukemia. (Funded by Sanofi-Aventis and Genentech; BCIRG-006 ClinicalTrials.gov number, NCT00021255.).

Maintaining normal hemoglobin levels with epoetin alfa in mainly nonanemic patients with metastatic breast cancer receiving first-line chemotherapy: a survival study.

PURPOSE: To evaluate the effect on survival and quality of life of maintaining hemoglobin (Hb) in the range of 12 to 14 g/dL with epoetin alfa versus placebo in women with metastatic breast cancer (MBC) receiving first-line chemotherapy. PATIENTS AND METHODS: Eligible patients were randomly assigned to receive epoetin alfa 40,000 U once weekly or placebo for 12 months. Study drug was initiated if baseline Hb was < or = 13 g/dL or when Hb decreased to < or = 13g/dL during the study. The primary end point was 12-month overall survival (OS). RESULTS: The study drug administration was stopped early in accordance with a recommendation from the Independent Data Monitoring Committee because of higher mortality in the group treated with epoetin alfa. Enrollment had been completed, with 939 patients enrolled (epoetin alfa, n = 469; placebo, n = 470). Most patients had Hb more than 12 g/dL at baseline (median Hb, 12.8 g/dL) or during the study. From the final analysis, 12-month OS was 70% for epoetin alfa recipients and 76% for placebo recipients (P = .01). Optimal tumor response and time to disease progression were similar between groups. The reason for the difference in mortality between groups could not be determined from additional subsequent analyses involving both study data and chart review. CONCLUSION: In this trial, the use of epoetin alfa to maintain high Hb targets in women with MBC, most of whom did not have anemia at the start of treatment, was associated with decreased survival. Additional research is required to clarify the potential impact of erythropoietic agents on survival when the Hb target range is 10 to 12 g/dL.

Capecitabine as adjuvant treatment for stage III colon cancer.

BACKGROUND: Intravenous bolus fluorouracil plus leucovorin is the standard adjuvant treatment for colon cancer. The oral fluoropyrimidine capecitabine is an established alternative to bolus fluorouracil plus leucovorin as first-line treatment for metastatic colorectal cancer. We evaluated capecitabine in the adjuvant setting. METHODS: We randomly assigned a total of 1987 patients with resected stage III colon cancer to receive either oral capecitabine (1004 patients) or bolus fluorouracil plus leucovorin (Mayo Clinic regimen; 983 patients) over a period of 24 weeks. The primary efficacy end point was at least equivalence in disease-free survival; the primary safety end point was the incidence of grade 3 or 4 toxic effects due to fluoropyrimidines. RESULTS: Disease-free survival in the capecitabine group was at least equivalent to that in the fluorouracil-plus-leucovorin group (in the intention-to-treat analysis, P<0.001 for the comparison of the upper limit of the hazard ratio with the noninferiority margin of 1.20). Capecitabine improved relapse-free survival (hazard ratio, 0.86; 95 percent confidence interval, 0.74 to 0.99; P=0.04) and was associated with significantly fewer adverse events than fluorouracil plus leucovorin (P<0.001). CONCLUSIONS: Oral capecitabine is an effective alternative to intravenous fluorouracil plus leucovorin in the adjuvant treatment of colon cancer.

Adjuvant docetaxel for node-positive breast cancer.

BACKGROUND: We compared docetaxel plus doxorubicin and cyclophosphamide (TAC) with fluorouracil plus doxorubicin and cyclophosphamide (FAC) as adjuvant chemotherapy for operable node-positive breast cancer. METHODS: We randomly assigned 1491 women with axillary node-positive breast cancer to six cycles of treatment with either TAC or FAC as adjuvant chemotherapy after surgery. The primary end point was disease-free survival. RESULTS: At a median follow-up of 55 months, the estimated rates of disease-free survival at five years were 75 percent among the 745 patients randomly assigned to receive TAC and 68 percent among the 746 randomly assigned to receive FAC, representing a 28 percent reduction in the risk of relapse (P=0.001) in the TAC group. The estimated rates of overall survival at five years were 87 percent and 81 percent, respectively. Treatment with TAC resulted in a 30 percent reduction in the risk of death (P=0.008). The incidence of grade 3 or 4 neutropenia was 65.5 percent in the TAC group and 49.3 percent in the FAC group (P<0.001); rates of febrile neutropenia were 24.7 percent and 2.5 percent, respectively (P<0.001). Grade 3 or 4 infections occurred in 3.9 percent of the patients who received TAC and 2.2 percent of those who received FAC (P=0.05); no deaths occurred as a result of infection. Two patients in each group died during treatment. Congestive heart failure and acute myeloid leukemia occurred in less than 2 percent of the patients in each group. Quality-of-life scores decreased during chemotherapy but returned to baseline levels after treatment. CONCLUSIONS: Adjuvant chemotherapy with TAC, as compared with FAC, significantly improves the rates of disease-free and overall survival among women with operable node-positive breast cancer.

Long-term efficacy and safety of zoledronic acid in the treatment of skeletal metastases in patients with nonsmall cell lung carcinoma and other solid tumors: a randomized, Phase III, double-blind, placebo-controlled trial.

BACKGROUND: The authors previously reported the efficacy of a dose of 4 mg of zoledronic acid in reducing skeletal complications in patients with bone metastases secondary to lung carcinoma and other solid tumors (except carcinomas of the breast and prostate). In the current study, they update these results and report the long-term efficacy and safety of 21 months of treatment with zoledronic acid in a randomized, placebo-controlled trial. METHODS: A total of 773 patients were randomized to receive zoledronic acid (4 mg or 8 mg) or placebo via a 15-minute infusion every 3 weeks for 21 months. The 8-mg dose later was reduced to 4 mg (8/4-mg group). The primary efficacy endpoint was the percentage of patients at 21 months with >/= 1 skeletal-related event (SRE) (pathologic fracture, spinal cord compression, radiation therapy to bone, or surgery to bone). Secondary analyses (time to first SRE, annual incidence of SREs, and multiple-event analysis) included hypercalcemia of malignancy. RESULTS: Fewer patients treated with zoledronic acid developed at least 1 SRE at 21 months compared with patients treated with placebo (39% of those treated at the 4-mg dose [P =0.127] and 36% of those treated at the 8/4-mg dose [P = 0.023], compared with 46% of those treated with placebo). Furthermore, 4 mg of zoledronic acid significantly delayed the median time to first SRE (236 days with 4 mg vs. 155 days with placebo; P = 0.009) and significantly reduced the annual incidence of SREs (1.74 per year with the 4-mg dose vs. 2.71 per year with placebo; P = 0.012). Moreover, the 4-mg dose of zoledronic acid was found to reduce the risk of developing a skeletal event by 31% (hazard ratio of 0.693; P = 0.003). Zoledronic acid was found to be well tolerated with long-term use; the most commonly reported adverse events in all treatment groups included bone pain and the transient, acute-phase reactions of nausea, anemia, and emesis. CONCLUSIONS: To the authors' knowledge, zoledronic acid is the first bisphosphonate to demonstrate long-term safety and efficacy in this patient population.