Intelligence Research Research Articles

Mathematical Frameworks for Threat Intelligence Analysis: Leveraging Graph Theory and Machine Learning for Cyber Threat Assessment

Because there are so many complex online risks, we need new ways to look at threat data. This study suggests a complete approach that combines graph theory and machine learning methods to make figuring out cyber threats better. The basic idea behind networks is graph theory, which lets us show the complicated connections between different things in a connected world. This approach gives a full picture of the danger scene by representing cyber entities and how they interact as nodes and lines in a graph. This makes it easier to spot trends and outliers. The system includes machine learning techniques that make use of the huge amount of data that is available for analyzing cyber threats. Supervised learning methods are used for classification tasks. These let threats be put into groups based on past data and known patterns of bad behavior. Unsupervised learning methods, on the other hand, make finding anomalies easier by noticing changes in how networks normally behave. These machine learning models learn to adapt to changing threats by being trained and improved over and over again. This makes methods for finding threats and stopping them more effective. Combining graph theory and machine learning makes it possible to get useful information from a huge number of different data sources. Graph-based analytics bring together different kinds of data, like network traffic, system logs, and threat intelligence feeds, into a single view. This helps you see the connections between things that don't seem to be related. Machine learning algorithms improve this analysis by finding small patterns and trends that point to bad behavior. This gives cybersecurity professionals the power to stop new threats before they happen. Scalability and freedom are built into the suggested system so it can adapt to changing cyber dangers and network platforms. It can handle big datasets and real-time streaming data well by using distributed computer structures and flexible machine learning methods. This makes sure that threats are found and dealt with quickly. Putting graph theory and machine learning together is a good way to make threat intelligence research better in defense.

Understand your shady neighborhood: An approach for detecting and investigating hacker communities

Cyber threat intelligence (CTI) researchers strive to uncover collaborations and emerging techniques within hacker networks. This study proposes an empirical approach to detect communities within hacker forums for CTI purposes. Eighteen algorithms are systematically evaluated, including state-of-the-art and benchmark methods for identifying overlapping and disjoint groups. Using discussions from five prominent English hacker forums, a comparative analysis examines the influence of the algorithms’ theoretical foundations on community detection. Since ground truths are unattainable for such networks, the study utilizes a multi-metric strategy, incorporating modularity, coverage, performance, and a newly introduced quality measure, Triplet Hub Potential, which quantifies the presence of influential hubs. The findings reveal that while modularity optimization algorithms such as Leiden and Louvain deliver consistent results, neighbor-based expanding techniques tend to provide superior performance. In particular, the Expansion algorithm stood out by uncovering granular hierarchical community structures. The ability to investigate these intimacies is helpful for CTI researchers. Ultimately, we suggest an approach to investigate hacker forums using community detection methods and encourage the future development of algorithms tailored to expose nuances within hacker networks.

OPPORTUNITIES OF USING ARTIFICIAL INTELLIGENCE IN SMALL AND MEDIUM-SIZED BUSINESSES

In the 21st century, artificial intelligence has been recognized as the main driver of universal progress. It has already gained a place in the economy of developed countries. It is used in both public and private sectors: in education, transport, finance and banking sector, medicine, and so on. Based on the scientific papers of artificial intelligence researchers, it is clear that small and medium-sized businesses can get significant benefits from artificial intelligence use. Small and medium-sized enterprises often work with limited resources, and automating repetitive tasks that can be performed with the help of artificial intelligence can help them save time and costs. Through artificial intelligence, SMEs are empowered to make data-driven decisions, improve their products or services, and identify new opportunities. The paper discusses the possibilities of using artificial intelligence in business, how AI can transform business operations and, therefore, the global economy, highlights the positive and negative aspects of AI, and expresses opinions in the direction of reducing the challenges associated with the introduction of artificial intelligence.

BCN20000: Dermoscopic Lesions in the Wild

Advancements in dermatological artificial intelligence research require high-quality and comprehensive datasets that mirror real-world clinical scenarios. We introduce a collection of 18,946 dermoscopic images spanning from 2010 to 2016, collated at the Hospital Clínic in Barcelona, Spain. The BCN20000 dataset aims to address the problem of unconstrained classification of dermoscopic images of skin cancer, including lesions in hard-to-diagnose locations such as those found in nails and mucosa, large lesions which do not fit in the aperture of the dermoscopy device, and hypo-pigmented lesions. Our dataset covers eight key diagnostic categories in dermoscopy, providing a diverse range of lesions for artificial intelligence model training. Furthermore, a ninth out-of-distribution (OOD) class is also present on the test set, comprised of lesions which could not be distinctively classified as any of the others. By providing a comprehensive collection of varied images, BCN20000 helps bridge the gap between the training data for machine learning models and the day-to-day practice of medical practitioners. Additionally, we present a set of baseline classifiers based on state-of-the-art neural networks, which can be extended by other researchers for further experimentation.

A Scientometric Review of Medical Artificial Intelligence Research: From A Social Science Perspective

A Scientometric Review of Medical Artificial Intelligence Research: From A Social Science Perspective

1987-LB: Limitations of Artificial Intelligence Research in Predicting Type 2 Diabetes Macrovascular Complications—A Scoping Review

1987-LB: Limitations of Artificial Intelligence Research in Predicting Type 2 Diabetes Macrovascular Complications—A Scoping Review

Gene selection based on recursive spider wasp optimizer guided by marine predators algorithm

Detecting tumors using gene analysis in microarray data is a critical area of research in artificial intelligence and bioinformatics. However, due to the large number of genes compared to observations, feature selection is a central process in microarray analysis. While various gene selection methods have been developed to select the most relevant genes, these methods’ efficiency and reliability can be improved. This paper proposes a new two-phase gene selection method that combines the ReliefF filter method with a novel version of the spider wasp optimizer (SWO) called RSWO-MPA. In the first phase, the ReliefF filter method is utilized to reduce the number of genes to a reasonable number. In the second phase, RSWO-MPA applies a recursive spider wasp optimizer guided by the marine predators algorithm (MPA) to select the most informative genes from the previously selected ones. The MPA is used in the initialization step of recursive SWO to narrow down the search space to the most relevant and accurate genes. The proposed RSWO-MPA has been implemented and validated through extensive experimentation using eight microarray gene expression datasets. The enhanced RSWO-MPA is compared with seven widely used and recently developed meta-heuristic algorithms, including Kepler optimization algorithm (KOA), marine predators algorithm (MPA), social ski-driver optimization (SSD), whale optimization algorithm (WOA), Harris hawks optimization (HHO), artificial bee colony (ABC) algorithm, and original SWO. The experimental results demonstrate that the developed method yields the highest accuracy, selects fewer features, and exhibits more stability than other compared algorithms and cutting-edge methods for all the datasets used. Specifically, it achieved an accuracy of 100.00%, 94.51%, 98.13%, 95.63%, 100.00%, 100.00%, 92.97%, and 100.00% for Yeoh, West, Chiaretti, Burcyznski, leukemia, ovarian cancer, central nervous system, and SRBCT datasets, respectively.

Machine Learning and Applications a Review Study

The proliferation of information and the construction of many databases have made the extraction of data from usable information a pressing issue that has to be addressed. Machine learning is the central focus of research in artificial intelligence. This paper provides an introduction to the definition of machine learning and its fundamental structure. Machine learning techniques including rote learning, inductive learning, explanatory learning, learning using neural networks, and knowledge discovery are also covered. In addition to outlining the present course of machine learning's progress, this article provides an introduction to its aims.

Preliminary study on AI-assisted diagnosis of bone remodeling in chronic maxillary sinusitis

ObjectiveTo construct the deep learning convolution neural network (CNN) model and machine learning support vector machine (SVM) model of bone remodeling of chronic maxillary sinusitis (CMS) based on CT image data to improve the accuracy of image diagnosis.MethodsMaxillary sinus CT data of 1000 samples in 500 patients from January 2018 to December 2021 in our hospital was collected. The first part is the establishment and testing of chronic maxillary sinusitis detection model by 461 images. The second part is the establishment and testing of the detection model of chronic maxillary sinusitis with bone remodeling by 802 images. The sensitivity, specificity and accuracy and area under the curve (AUC) value of the test set were recorded, respectively.ResultsPreliminary application results of CT based AI in the diagnosis of chronic maxillary sinusitis and bone remodeling. The sensitivity, specificity and accuracy of the test set of 93 samples of CMS, were 0.9796, 0.8636 and 0.9247, respectively. Simultaneously, the value of AUC was 0.94. And the sensitivity, specificity and accuracy of the test set of 161 samples of CMS with bone remodeling were 0.7353, 0.9685 and 0.9193, respectively. Simultaneously, the value of AUC was 0.89.ConclusionIt is feasible to use artificial intelligence research methods such as deep learning and machine learning to automatically identify CMS and bone remodeling in MSCT images of paranasal sinuses, which is helpful to standardize imaging diagnosis and meet the needs of clinical application.

Combined with the semantic features of CT and selected clinical variables, a machine learning model for accurately predicting the prognosis of omicron was established

Abstract Objective To efficiently use medical resources and offer optimal personalized treatment for individuals with Omicron infection, it's vital to predict the disease's outcome early on. This research developed three machine learning models to foresee the results for Omicron-infected patients. Methods Data from 253 Omicron-infected patients, including their CT scans, clinical details, and relevant laboratory values, were studied. The patients were categorized into two groups based on their disease progression: favorable prognosis and unfavorable prognosis. Patients manifesting respiratory failure, acute liver or kidney impairment, or fatalities were placed in the “poor” group. Those lacking such symptoms were allocated to the “good” group. The participants were randomly split into training set (202) and validation set (51) with an 8:2 ratio. Radiomics features were produced using image processing, focused segmentation, feature extraction, and selection, leading to the establishment of a radiomics model. A univariate logistic regression method identified potential clinical factors contributing to a clinical model's development. Eventually, the fused feature set, integrating radiomics features and clinical indicators, was used for the combined model. The model's prediction performance was assessed using the area under the receiver operating characteristic curve (AUC). The model's clinical usefulness was evaluated by generating calibration and decision curves. Results Compared to other classification models, the combined model showcased the best classification performance. It achieved an AUC of 0.848 and accuracy of 0.763 in the training set, and 0.797 and 0.750 in the validation set, respectively. Conclusion This study employed machine learning model to accurately predict the prognosis of Omicron-infected patients. ADVANCES IN KNOWLEDGE： (1) Topic innovation: At present, there is a lack of research on the use of CT images to construct machine learning models to predict the prognosis of patients with Omicjon infection. This study intends to establish clinical, radiomics and combined models to provide more possibilities for the identification of the two. (2) Platform innovation: The feature extraction and screening and the establishment of omics model in this study will be completed in the intelligent scientific research platform, which can reduce the error caused by human error, simplify the operation steps and save the time of data processing time.

Understanding AI innovation contexts: a review and content analysis of artificial intelligence and entrepreneurial ecosystems research

PurposeAn emerging research stream focuses on the place-based ecosystems where artificial intelligence (AI) innovations emerge and develop. This literature builds on the contextual turn in management research and, specifically, work on entrepreneurial ecosystems. However, as a nascent research area, the literature on AI and entrepreneurial ecosystems is fragmented across academic and practitioner boundaries and unconnected disciplines because of disparate and ill-defined concepts. As a result, the literature is disorganized and its main insights are latent. The purpose of this paper is to synthesize research on AI ecosystems and identify the main insights.Design/methodology/approachWe first consolidate research on the “where” of AI innovation through a scoping review. To address the fragmentation in the literature and understand how entrepreneurial ecosystems are associated with AI innovation, we then use content analysis to explore the literature.FindingsWe identify the main characteristics of the AI and ecosystems literature and the key dimensions of “AI entrepreneurial ecosystems”: the local actors and factors in geographic territories that are coordinated to support the creation and development of AI technologies. We clarify the relationships among AI technologies and ecosystem dimensions and uncover the latent themes and underlying structure of research on AI entrepreneurial ecosystems.Originality/valueWe increase conceptual precision by introducing and defining an umbrella concept—AI entrepreneurial ecosystem—and propose a research agenda to spur further insights. Our analysis contributes to research at the intersection of management, information systems, and entrepreneurship and creates actionable insights for practitioners influenced by the geographic agglomeration of AI innovation.

Integrasi Kecerdasan Buatan Dalam Sistem Rekomendasi Produk Untuk E-Commerce

The increasing popularity of e-commerce has driven the development of more sophisticated product recommendation systems to enhance user experience. Within this framework, the incorporation of Artificial Intelligence (AI) has become essential for enhancing the caliber of product suggestions through the examination of user behavioral patterns and product attributes. This paper explores the incorporation of artificial intelligence into e-commerce product recommendation systems. This strategy integrates AI methodologies like machine learning, data mining, and natural language processing (NLP) to produce recommendation systems that are personalized and pertinent. Through in-depth analysis of user data and product information, the system can predict user preferences with higher accuracy, improve user engagement levels, and ultimately increase sales conversions. By combining domain knowledge, technical expertise, and the latest research methodologies, This research offers valuable insights for both e-commerce system developers and artificial intelligence researchers. The integration of artificial intelligence into product recommendation systems represents a critical measure in facilitating growth and fostering innovation within the ever-changing e-commerce landscape.

Long-term Treatment Patterns for Diabetic Macular Edema: Up to 6-Year Follow-up in the IRIS® Registry

To characterize anti-vascular endothelial growth factor (VEGF) intravitreal therapy (IVT) patterns and long-term visual outcomes among patients with diabetic macular edema (DME) in routine clinical practice in the United States. Retrospective analysis of the American Academy of Ophthalmology's Intelligent Research in Sight (IRIS®) Registry.; Participants: Treatment-naïve patients with DME (no previous IVT in the past 12 months) initiating anti-VEGF IVT from 1/1/2015-3/31/2021. Baseline characteristics, treatment patterns, and long-term visual acuity (VA) outcomes were reported for up to 6 years of follow-up. Outcomes included the annualized number of injections, change in VA, and anti-VEGF agents. A total of 190,345 eyes met inclusion criteria. After 1 year of anti-VEGF IVT initiation, eyes received a mean of 3.9 (±2.8) injections and gained +3.2 (±16.4) letters of vision. Of the 1,236 eyes with year 6 data, eyes received a mean of 2.9 (±2.1) injections in year 6 and gained +0.5 (±19.7) letters from baseline. The number of injections decreased, and injection intervals increased year over year up to 6 years regardless of baseline VA initiation. The average injection interval was 10-weeks in year 1, then widened to 13.2 in year 2, before plateauing in years 3-6 (12.6, 12.3, 12.2, and 12.3 weeks respectively). Improvements in VA from baseline were greatest in eyes that received 5 or more injections each year. At the end of follow-up, eyes with good baseline vision (> 20/25) lost vision, whereas those with worse baseline vision (< 20/25) gained vision. Although 51.7% of patients with DME discontinued IVT after a mean of 6 months, 32.8% re-initiated anti-VEGF IVT. Worse VA outcomes were associated with patients of Hispanic ethnicity (-1.08 [-1.34, -0.83] compared to non-Hispanic), Medicaid insurance (-1.15 [-1.48, -0.81] compared to Commercial), and older age (-0.06 [-0.07, -0.05] each additional year) CONCLUSIONS: Patients with DME in the routine clinical settings receive fewer injections than those in clinical trials and fewer than recommended per the label of FDA approved anti-VEGF IVT.

A bibliometric analysis of the advance of artificial intelligence in medicine

This bibliometric study analyzes the evolution of research in artificial intelligence (AI) applied to medicine from 2015 to September 2023. Using the Scopus database and keywords related to AI, machine learning, and deep learning in medicine, tools such as VOSviewer and Bibliometrix were used to explore publication trends, subject areas, co-authorship networks, and the most productive countries, among others. 2,064 articles were analyzed, and a significant increase in global academic production has been evident in the last five years. International collaboration was notable, with China and the United States leading in knowledge contribution. The keyword analysis highlights the breadth of topics and applications of AI in medicine, with particular emphasis on cancer detection, dengue diagnosis, and medical image analysis, among others. In conclusion, this study highlights the growing academic interest in the application of AI in medicine and the need for collaborative research. The findings underscore the relevance of these technologies in key areas of health care, contributing significantly to advances in medical diagnosis and prognosis.

Conceptualizing understanding in explainable artificial intelligence (XAI): an abilities-based approach

A central goal of research in explainable artificial intelligence (XAI) is to facilitate human understanding. However, understanding is an elusive concept that is difficult to target. In this paper, we argue that a useful way to conceptualize understanding within the realm of XAI is via certain human abilities. We present four criteria for a useful conceptualization of understanding in XAI and show that these are fulfilled by an abilities-based approach: First, thinking about understanding in terms of specific abilities is motivated by research from numerous disciplines involved in XAI. Second, an abilities-based approach is highly versatile and can capture different forms of understanding important in XAI application contexts. Third, abilities can be operationalized for empirical studies. Fourth, abilities can be used to clarify the link between explainability, understanding, and societal desiderata concerning AI, like fairness and trustworthiness. Conceptualizing understanding as abilities can therefore support interdisciplinary collaboration among XAI researchers, provide practical benefit across diverse XAI application contexts, facilitate the development and evaluation of explainability approaches, and contribute to satisfying the societal desiderata of different stakeholders concerning AI systems.

An intelligent research environment on cotton diseases and pests based on a cotton phytosanitary surveillance ontology ontoSYSPARCOTCI.

User-centred systems need to incorporate solutions to make them easier to use. Based on a domain ontology, our system has an extensive knowledge base that is populated by a semantic web platform that we have built. This knowledge base needs to be queried on demand by users due to the limitations of the search engine integrated into the semantic web platform. However, this task is not trivial for a human being due to the complexity of the schemas making up the knowledge base and the inability to perform SPARQL queries to obtain answers. In this paper, we propose an approach based on skills questions that we have defined with domain experts and that have helped validate our domain ontology. This approach allows users to ask their questions in natural language to obtain an answer. To achieve our goal, we used natural language processing tools, in particular Spacy's pre-trained French pipeline. The TF-IDF weighting method was used to calculate the similarity score between two sentences.

Artificial Intelligence: A Quick Overview of Speech Recognition System

Abstract: The study of intelligent behaviour is known as artificial intelligence. Wagner said, "We build intelligent machines in science." By this, he implied that the primary goal is to teach these sophisticated devices, such as computers, to examine and comprehend human conduct. numerous believed that human intelligence might be understood through the construction of several programmes, while other researchers believed that numerous basics should be required in order to accomplish the intended objective. We can carry out or replicate any type of task we conduct using a computer. Therefore, computers are the ideal devices for developing artificial intelligence. In the hopes that they will outperform computers in several other areas, some researchers created additional computing devices. Billions of dollars were invested in creating new devices that surpassed the speed of the computer used to simulate software. However, the computers and the code both need to be extremely quick. As you learn about artificial intelligence, you may wonder, what really is intelligence? How is intelligence measured? Turn become something quite significant. In this case, the programme that evaluates its surroundings acts as the intelligent agent. Modern digital computers and human minds are comparable to one other in ways such as symbolic information processing systems. Both use symbolic data as input, change it in accordance with predetermined guidelines, and then use the results to solve issues. In order to mechanically replicate intelligent human activity, artificial intelligence researchers have programmed some algorithms to recognise intelligent human behaviour. Human conduct in language processing, chess, and other games, as well as medical diagnostics, may be used to monitor this activity. Voice recognition is a key component of AI. A certain voice may be recognised using voice recognition technology. The cornerstone for speaker identification is voice signals. Voice targeting is applicable to a wide range of applications, including voice mail, database access, phone banking, and phone purchasing. The ability to enter one's voice for verification is among the most potent uses of voice recognition for security. The fundamental means of interpersonal communication is speech. The technique of translating voice sounds into appropriate text is known as speech recognition. Over the past few years, speech recognition technology has advanced significantly. Nonetheless, there are other significant study obstacles, such as variations in speaker and language, ambient sound, word size, etc. This study aims to give a comprehensive overview of speech acceptance by summarising the several approaches utilised in the standard speech system and describing the numerous processes involved

Evolving Client–Vendor Relationship: As Manifested Through an Artificial Intelligence Research Unit of an Indian IT Organization

The Indian information technology (IT) sector is traditionally known for delivering IT and business process management (BPM) services to clients across the globe. To ensure the longevity of this client and service provider relationship, Indian IT organizations have made significant client-specific investments and actively engaged them in offshore development activities. One of the outcomes from such investments can be seen in terms of the very high maturity standards that Indian IT organizations currently exhibit in the context of IT–BPM service delivery. In recent times, these organizations are also undertaking offshore projects around emerging technologies. Such projects, especially those related to artificial intelligence (AI), are largely centred around the automation of client-focused BPM services. Given their relative nascency, such projects are primarily led by the AI research units that are rapidly diffusing across Indian IT organizations. In this study, we rely on a four-month-long ethnography to explore a client and vendor (or service-provider) relationship as manifested through one such AI research unit of an Indian IT organization situated in Bengaluru, India. Our objective from this study is to understand how these new offshore AI projects affect the longevity of the relationship between clients and the Indian IT service providers. Given the nature of emerging technology workflows, our findings indicate a tilt more towards meeting client-side expectations than that of vendors in these projects. We portend that this can potentially disrupt the longevity of the client–vendor relationship in the Indian IT services sector, and, therefore, both the clients and vendors must attend to the peculiarities of offshore emerging technology projects to mitigate such disruptions.

P.074 Assessing the emergence and evolution of artificial intelligence and machine learning research in neuroradiology

Background: Interest in artificial intelligence (AI) and machine learning (ML) has been growing in neuroradiology, but there is limited knowledge on how this interest has manifested into research and the field’s trends, challenges, and future directions. Methods: The American Journal of Neuroradiology was queried for original research articles published since inception (Jan. 1, 1980) to Sept. 19, 2022 that contained any of the following key terms: “machine learning”, “artificial intelligence”, or “radiomics”. Articles were screened, categorized into Statistical Modelling (Type 1), AI/ML Development (Type 2), or End-user Application (Type 3) and then bibliometrically analyzed. Results: A total of 124 articles were identified with 85% being non-integration focused (Type 1 n = 41, Type 2 n = 65) and the remaining (n = 18) being Type 3. The total number of articles published grew two-fold in the last five years, with Type 2 articles mainly driving this growth. While most (66%) Type 2 articles were led by a radiologist with 55% possessing a postgraduate degree, a minority of Type 2 articles addressed bias (15%) and explainability (20%). Conclusions: The results of this study highlight areas for improvement but also strengths that stakeholders can consider when promoting the shift towards integrating practical AI/ML solutions in neuroradiology.

Construction of Accounting Fraud and Its Audit Countermeasure Model Based on Computer Technology

A prediction model of financial fraud of listed companies based on machine learning method is proposed to predict financial fraud of listed companies. Using the data set of Chinese listed companies from 2000 to 2020 as observation samples, Benford’s Law, LOF local anomaly method and SMOTE oversample were adopted, grey samples were excluded, and characteristic variables were selected from five aspects: fraud motivation, solvency, profitability, cash flow and operating capacity. The financial fraud identification model Xscore is established based on the XGBoost method. The Xscore model can improve the accuracy of model prediction, and is superior to the Fscore model and Cscore model in accuracy, recall rate, AUC index, KS value, PSI stability, etc. It is more suitable for predicting the financial fraud of listed companies in China. The results of this study are helpful in promoting the research and application of artificial intelligence and machine learning in accounting, and provide references for promoting the disclosure of high-quality financial information by listed companies and maintaining the order of the capital market.