| viXra.org > Artificial Intelligence > 2406 |
 |
 |
| viXra.org > Artificial Intelligence > 2406 |
|
|
[14] viXra:2406.0170 [pdf] submitted on 2024-06-28 20:50:32
Authors: Hui Liu
Comments: 4 Pages. (Note by viXra Admin: Please cite and list scientific references)
This paper explores the basic composition and operational mechanisms of intelligent systems. Intelligence is defined as the ability to solve problems, and the operation of intelligent systems is centered around databases. The three fundamental elements of intelligent system operation include the construction, retrieval, and use of databases. This paper discusses in detail the process of handling a single event in a single thread. Complex event composites can be broken down into multiple single events for resolution.
Category: Artificial Intelligence
[13] viXra:2406.0166 [pdf] submitted on 2024-06-28 20:44:48
Authors: Tanvir Rahman, Ataur Rahman, Tamanna Afroz
Comments: 6 Pages.
The major player in the revolution of early detection and diagnosis of brain tumors, with great implications for patient outcomes, is medical image processing. It is an inherently difficult and time-consuming task to manually classify brain tumors by experienced experts, even though manual classification has proven effective. A promising avenue has emerged as the integration of automatic segmentation techniques, which promises improved efficiency and performance in response to these challenges. This long work aims to provide an in-depth and critical analysis of MRI-based brain tumor segmentation techniques, with a critical eye toward the most recent developments in automatic segmentation techniques. Our analysis explores the rapidly changing field of completely automatic segmentation approaches, which diverges from the evaluations that mostly focus on traditional methodologies. The discussion opens with a broad summary that emphasizes how important brain tumor segmentation is to medical image processing as a whole. Here, we highlight how crucial precise segmentation is to facilitating early detection and guiding treatment choices later on. We recognize the difficulties that come with manual segmentation procedures and explain why automation segmentation techniques are necessary to reduce these difficulties and bring about increased productivity. The central section of the work navigates the complex terrain of cutting-edge algorithms, enabling a thorough investigation of the most recent developments in completely autonomous segmentation techniques. This thorough explanation highlights the growing acceptance and increased effectiveness of modern methods while addressing the complexities and difficulties present in the field of brain tumor segmentation. Using specially crafted neural networks, our research is unique in that it concentrates on the paradigm shift toward fully autonomous segmentation. Brain tumor segmentation has been transformed by the incorporation of deep learning techniques, which enable complex pattern recognition and nuanced analysis using medical imaging data. Our efforts have resulted in the creation of a unique neural network model specifically intended for the automated identification of brain malignancies. The talk highlights how deep learning techniques can have a revolutionary effect, and it ends with the creation of a sophisticated custom neural network model. Our model demonstrates its ability to accurately and automatically detect brain tumor boundaries by achieving a remarkable level of accuracy.
Category: Artificial Intelligence
[12] viXra:2406.0165 [pdf] submitted on 2024-06-28 17:36:46
Authors: Tanvir Rahman
Comments: 5 Pages.
Monkeypox is a viral disease that affects bothanimals and humans. Monkeypox can have a substantial negative influence on human health, particularly in areas with a lack of healthcare services. The sickness can produce epidemics, and it might be difficult to stop the spread of the disease. For effective treatment and to stop the disease from spreading further, early identification and detection of monkeypox are essential. Therefore, the healthcare industrymay benefit from the development of precise and effective methods for the detection of monkeypox, such as image classification. In this paper, we propose a novel approach for detecting Monkeypox using image classification. The proposed method utilizes a Transfer Learning Model and other machine learning models to classify images of patients with Monkeypox.The system employs a majority voting technique to improve the accuracy of the classification. The proposed system is evaluated using a dataset of images obtained from patients withMonkeypox, and the results show that the proposed approach achieves high accuracy in detecting Monkeypox. The proposed system has the potential to assist healthcare professionals indiagnosing and treating patients with Monkeypox, and it can contribute to the efforts of controlling the spread of the disease
Category: Artificial Intelligence
[11] viXra:2406.0161 [pdf] replaced on 2024-08-03 15:24:09
Authors: Ait-Taleb nabil
Comments: 5 Pages.
In this article, we will describe the mechanism that links the notion of causality to correlations. This article answers yes to the following question: Can we deduce a causal relationship from correlations?
Category: Artificial Intelligence
[10] viXra:2406.0156 [pdf] submitted on 2024-06-26 19:18:42
Authors: Junhao Yu, Fuyuan Xiao
Comments: 2 Pages. (Note by viXra Admin: Please cite and list scientific references)
In this paper, a novel complex dual Gaussian fuzzy number (CDGFN) is proposed to more accurately model two-dimensional uncertainty, which serves as the medium to represent generalized quantum basic belief assignment (GQBBA).
Category: Artificial Intelligence
[9] viXra:2406.0075 [pdf] submitted on 2024-06-15 17:56:44
Authors: Agnij Moitra
Comments: 16 Pages.
Gradient boosting is a widely used machine learning algorithm for tabular regression, classification and ranking. Although, most of the open source implementations of gradient boosting such as XGBoost, LightGBM and others have used decision trees as the sole base estimator for gradient boosting. This paper, for the first time, takes an alternative path of not just relying on a static base estimator (usually decision tree), and rather trains a list of models in parallel on the residual errors of the previous layer and then selects the model with the least validation error as the base estimator for a particular layer. This paper has achieved state-of-the-art results when compared to other gradient boosting implementations on 50+ tabular regression and classification datasets. Furthermore, ablation studies show that MSBoost is particularly effective for small and noisy datasets. Thereby, it has a significant social impact especially in tabular machine learning problems in the domains where it is not feasible to obtain large high quality datasets.
Category: Artificial Intelligence
[8] viXra:2406.0056 [pdf] submitted on 2024-06-11 21:32:40
Authors: Philip Naveen
Comments: 42 Pages.
This manuscript is merely a formal documentation of the purpose and details surrounding the online convex optimization toolbox (OCOBox) for MATLAB. The purpose of this toolbox is to provide a collection of algorithms that work under stochastic situations where traditional algorithmic theory does not fare so well. The toolbox encompasses a wide range of methods including Bayesian persuasion, bandit optimization, Blackwell approachability, boosting, game theory, projection-free algorithms, and regularization. In the future, we plan to extend OCOBox to interactive machine learning algorithms and develop a more robust GUI.
Category: Artificial Intelligence
[7] viXra:2406.0037 [pdf] submitted on 2024-06-08 04:51:00
Authors: Fuyuan Xiao
Comments: 3 Pages.
In this paper, we propose a quantum evidential reasoning rule in the framework of generalized quantum evidence theory.
Category: Artificial Intelligence
[6] viXra:2406.0035 [pdf] submitted on 2024-06-07 01:17:32
Authors: Junjie Huang, Fuyuan Xiao
Comments: 1 Page.
In this paper, to extend the triditional evidential reasoning (ER) method to complex plane, a novelcomplex evidential reasoning (CER) method is defined in the framework of complex evidencetheory (CET).
Category: Artificial Intelligence
[5] viXra:2406.0012 [pdf] submitted on 2024-06-03 21:03:31
Authors: Taeho Jo
Comments: 13 Pages.
This article proposes the modified KNN (K Nearest Neighbor) algorithm which receives a graph as its input data and is applied tothe text summarization. The graph is more graphical for representing a word and the text summarization is able to be viewed into a binaryclassification where each paragraph is classified into summary or non-summary. In the proposed system, a text which is given as theinput is partitioned into a list of paragraphs, each paragraph is classified by the proposed KNN version, and the paragraphs which areclassified into summary are extracted ad the output. The proposed KNN version is empirically validated as the better approach in deciding whether each paragraph is essential or not in news articles and opinions. In this article, a paragraph is encoded into a weighted and undirected graph and it is represented into a list of edges.
Category: Artificial Intelligence
[4] viXra:2406.0011 [pdf] submitted on 2024-06-03 21:03:18
Authors: Taeho Jo
Comments: 13 Pages.
This article proposes the modified KNN (K Nearest Neighbor) algorithm which considers the feature similarity and is applied to the text segmentation. The words which are given as features for encoding words into numerical vectors have their own meanings and semantic relations with others, and the text segmentation is able to be viewed into a binary classification where each adjacent paragraphpair is classified into boundary or continuance. In the proposed system, a list of adjacent paragraph pairs is generated by sliding atext with the two sized window, each pair is classified by the proposed KNN version, and the boundary is put between the pairs which are classified into boundary. The proposed KNN version is empirically validated as the better approach in deciding whether each pair should be separated from each other or not in newsarticles and opinions. The significance of this research is to improve the classification performance by utilizing the feature similarities.
Category: Artificial Intelligence
[3] viXra:2406.0010 [pdf] submitted on 2024-06-03 21:02:49
Authors: Taeho Jo
Comments: 12 Pages.
This article proposes the modified KNN (K Nearest Neighbor) algorithm which receives a string vector as its input data and isapplied to the text segmentation. The results from applying the string vector based algorithms to the text categorizations were successful in previous works, and the text segmentation is able to be viewed into a binary classification where each adjacent paragraph pair is classified into boundary or continuance. In the proposedsystem, a list of adjacent paragraph pairs is generated by sliding a text with the two sized window, each pair is classified by theproposed KNN version, and the boundary is put between the pairs which are classified into boundary. The proposed KNN version isempirically validated as the better approach in deciding whether each pair should be separated from each other or not in news articles and opinions. We need to define and characterizemathematically more operations on string vectors for modifying more advanced machine learning algorithms.
Category: Artificial Intelligence
[2] viXra:2406.0009 [pdf] submitted on 2024-06-03 21:02:38
Authors: Taeho Jo
Comments: 12 Pages.
This article proposes the modified KNN (K Nearest Neighbor) algorithm which receives a graph as its input data and is applied tothe text segmentation. The graph is more graphical for representing a word and the text segmentation is able to be viewed into a binaryclassification where each adjacent paragraph pair is classified into boundary or continuance. In the proposed system, a list of adjacentparagraph pairs is generated by sliding a text with the two sized window, each pair is classified by the proposed KNN version, and theboundary is put between the pairs which are classified into boundary. The proposed KNN version is empirically validated as thebetter approach in deciding whether each pair should be separated from each other or not in news articles and opinions. In this article, an adjacent paragraph pair is encoded into a weighted and undirected graph and it is represented into a list of edges.
Category: Artificial Intelligence
[1] viXra:2406.0001 [pdf] submitted on 2024-06-01 18:57:25
Authors: Vansh Kumar
Comments: 16 Pages.
This paper introduces Vision, a novel 175-billion parameter multimodal AI model.Vision is trained from scratch to natively understand text, images, video, and audioand to generate text and images, setting it apart from existing models. Developedwith a focus on incorporating Indian context, values, and culture, Vision aims to em-power users with a culturally relevant AI experience. A unique security feature allowsgenerated images to be backtracked to Vision, mitigating concerns about potential mis-use for misinformation. Evaluations on standard benchmarks demonstrate that Visionachieves state-of-the-art performance in a diverse range of tasks, including reasoning,solving mathematical problems, code generation, and image understanding. Further-more, Vision exhibits remarkable proficiency in multilingual chat, supporting a widearray of global languages as well as regional Indian languages such as Hindi, Punjabi,and Marathi. We believe that Vision represents a significant step towards buildingmore inclusive and culturally relevant AI systems, with the potential to positively im-pact various domains in India and beyond.
Category: Artificial Intelligence
| Third-Party Links: |
|