| viXra.org > Artificial Intelligence > 2006 |
 |
 |
| viXra.org > Artificial Intelligence > 2006 |
|
|
[19] viXra:2006.0265 [pdf] submitted on 2020-06-29 13:57:50
Authors: Samuel Kopelowitz, Uday Reddy
Comments: 25 Pages.
Twitter data mining techniques have been used in the run-up to elections to predict their outcomes and perform analysis to explain results. Due to the popularity of the social media platform it is possible to collect large amounts of data with which often lexicon-based sentiment analysis has been used to accomplish these tasks, mostly because of its efficiency and simplicity. More recently, hybrid techniques, which in addition to calculating tweet sentiment also incorporate topic modelling methods to extract the main “topics” from a corpus of text, have been applied independently for both election prediction and analysis. It is possible to use hybrid methods to analyse different political issues (e.g. economic, social, etc) and the public opinion for candidates in respect to them; and other hybrid methods have been shown to outperform baseline sentiment analysis approaches for election prediction. A mining solution which can accomplish both of these tasks non-exhaustively is desirable for better predictions and a greater understanding of election outcomes. This report will present a novel approach to mining Twitter data, Hybrid Topic-Based Sentiment Analysis with Issue Filtering (HTBSA*), which will not only pose as a potential improvement upon state-of-the-art techniques for election prediction; but can be abstracted to perform candidate analysis on any individual political issue, proposing a baseline methodology for doing this. This research approach has effectively outperformed all of the well-established methods in the realm of lexicon-based election prediction, giving a mean average error as low as 2.20% from true vote share. This technique was performed on data collected on the run up to the UK General Election 2019 and in an addition to this, it has successfully been black box tested on an unseen dataset. Based on the empirical evidence given by our results, HTBSA* can be relied upon to predict elections occurring in the future, but analysis results in respect to individual political issues may be inconsistent, suggesting further work is required. Lines of research that come as a result of this study have the potential to tackle election mining problems in new ways, which are more sophisticated than what has been done previously.
Category: Artificial Intelligence
[18] viXra:2006.0237 [pdf] submitted on 2020-06-26 07:25:51
Authors: Andrei P. Kirilyuk
Comments: 67 pages, 43 eqs, 86 refs
While practical efforts in the field of artificial intelligence grow exponentially, the truly scientific and mathematically exact understanding of the underlying phenomena of intelligence and consciousness is still missing in the conventional science framework. The inevitably dominating empirical, trial-and-error approach has vanishing efficiency for those extremely complicated phenomena, ending up in fundamentally limited imitations of intelligent behaviour. We provide the first-principle analysis of unreduced many-body interaction process in the brain revealing its qualitatively new features, which give rise to rigorously defined chaotic, noncomputable, intelligent and conscious behaviour. Based on the obtained universal concepts of unreduced dynamic complexity, intelligence and consciousness, we derive the universal laws of intelligence applicable to any kind of intelligent system interacting with the environment. We finally show why and how these fundamentally substantiated and therefore practically efficient laws of intelligent system dynamics are indispensable for correct AI design and training, which is urgently needed in this time of critical global change towards the truly sustainable development.
Category: Artificial Intelligence
[17] viXra:2006.0235 [pdf] submitted on 2020-06-25 11:22:05
Authors: Yige Xue, Yong Deng
Comments: 15 Pages.
Mass function vector is used to handle uncertainty. Quaternion number is the extent of real number. The mass function vector can extend the mass function by combining the vector. In this paper, the mass function vector is extended by quaternion number, named as Quaternion Mass Function Vector(QMFV). The proposed QMFV has the advantage to deal with uncertain information. When the quaternion number degenerates into the real number, then the QMFV degenerates into the quaternion mass function. In addition, if the probability of multiple subsets of frame of discernment is not assigned to the single subsets, then the mass function vector will degenerate into mass function in classical evidence theory. When the quaternion number degenerates into the real number, then the combination rule of quaternion mass function vectors degenerates into the combination rule of mass function vectors. In the case when the probability of multiple subsets of frame of discernment is not assigned to the single subsets, the combination rule of mass function vectors degenerates into generalized dempster's rule of combination.
Numerical examples are applied to prove the efficiency of the proposed model. The experimental results show that the proposed model can apply the quaternion theory to mass function vector effectively and successfully.
Category: Artificial Intelligence
[16] viXra:2006.0210 [pdf] submitted on 2020-06-22 22:30:46
Authors: Yong Deng
Comments: 17 Pages.
Mass function is used to handle uncertainty. Quaternion number is the extent of imaginary number. In this paper, the classical mass function is extended by quaternion number, named as Quaternion Mass Function (QMF). The proposed QMF has the advantage to deal with uncertain information. When the quaternion number degenerates into the complex number, then the QMF degenerates into the complex mass function. In addition, if the complex mass function is degenerated as real number, the QMF is the same as mass function in classical evidence theory. In the case when the quaternion number degenerates into the real number and the QMF focus on the frame of discernment with single subsets, the QMF is the same as the probability distribution in probability theory. The combination rule is also presented to combine two QMFs, which is the generalization of Dempster rule. In the case when the quaternion mass function degenerates into the real number and assigns only to single subsets, the proposed combination rule is degenerated as Beyesian updation in probability theory. Numerical examples are applied to prove the efficiency of the proposed model. The experimental results show that the proposed model can apply the quaternion theory to mass function effectively and successfully.
Category: Artificial Intelligence
[15] viXra:2006.0208 [pdf] replaced on 2020-12-10 03:50:56
Authors: Jeongik Cho
Comments: 17 Pages.
Finding a latent code that can generate specific data by inverting a generative model is called latent code recovery (or latent vector recovery). When performing gradient descent based latent recovery, the probability that the recovered latent code was sampled from a latent random variable can be very low. To prevent this, latent regulation losses or element resampling methods have been used in some papers.
In this paper, when the latent random variable is an IID (Independent and Identically Distributed) random variable and performing gradient descent-based latent code recovery, we propose statistical distance latent regulation loss to maximize the probability that the latent code was sampled from the latent random variable. The statistical distance latent regulation loss is the distance between the discrete uniform distribution, assuming each latent code element has the same probability and one-dimensional distribution that each element of the latent random variable follows in common. Since the statistical distance latent regulation loss considers all elements simultaneously, it maximizes the probability that the latent code was sampled from a latent random variable.
Also, we propose the latent distribution goodness of fit test, an additional test that verifies whether the latent code is sampled from the latent random variable. This additional test verifies whether all recovered latent codes’ elements’ distribution follows one-dimensional distribution that each element of the latent random variable follows in common when the latent random variable is an IID random variable. Passing the latent distribution goodness of fit test does not mean that the latent codes are recovered correctly, but when the latent codes are recovered correctly, the latent distribution goodness of fit test should be passed.
Compared with other latent regulation losses or element resampling methods, only latent code recovery using the statistical distance latent regulation loss could recover the correct latent code with high performance in the gradient descent-based latent code recovery.
Category: Artificial Intelligence
[14] viXra:2006.0196 [pdf] submitted on 2020-06-20 21:40:35
Authors: Shaif Chowdhury, Soummyopriyo Chattopdhyay, Tapan Kumar Hazra
Comments: 10 Pages.
In this Paper, we are presenting a traffic surveillance system for detection and classification of vehicles in large scale videos. Vehicle detection is crucial part of Road safety. There are lots of different intelligent systems proposed for traffic surveillance. The system presented here is based on two steps, a descriptor of the image type haar-like, and a classifier type convolutional neural networks. A cascade classifier is used to extract objects rapidly and a neural network is used for final classification of cars. In case of Haar Cascades, the learning of the system is performed on a set of positive images (vehicles) and negative images (non-vehicle), and the test is done on another set of scenes. For the second, we have used faster R-CNN architecture. The cascade classifier gives faster processing time and Neural Network is used to increase the detection rate.
Category: Artificial Intelligence
[13] viXra:2006.0159 [pdf] submitted on 2020-06-18 06:23:41
Authors: Nazakat Ali
Comments: 10 Pages.
Chatbot is a technology that is used to mimic human behavior using natural language. There are different types of Chatbot that can be used as conversational agent in various business domains in order to increase the customer service and satisfaction. For any business domain, it requires a knowledge base to be built for that domain and design an information retrieval based system that can respond the user with a piece of documentation or generated sentences. The core component of a Chatbot is Natural Language Understanding (NLU) which has been impressively improved by deep learning methods. But we often lack such properly built NLU modules and requires more time to build it from scratch for high quality conversations. This may encourage fresh learners to build a Chatbot from scratch with simple architecture and using small dataset, although it may have reduced functionality, rather than building high quality data driven methods. This research focuses on Named Entity Recognition (NER) and Intent Classification models which can be integrated into NLU service of a Chatbot. Named entities will be inserted manually in the knowledge base and automatically detected in a given sentence. The NER model in the proposed architecture is based on artificial neural network which is trained on manually created entities and evaluated using CoNLL-2003 dataset.
Category: Artificial Intelligence
[12] viXra:2006.0126 [pdf] submitted on 2020-06-14 13:57:01
Authors: Davide Zagami
Comments: 5 Pages.
We provide a rigorous analysis of AIXI's behaviour under repeated Newcomblike settings. In this context, a Newcomblike problem is a setting where an agent is tied against an environment that contains a perfect predictor, whose predictions are used to determine the environmet's outputs. Since AIXI lacks good convergence properties, we chose to focus the analysis on determining whether an environment appears computable to AIXI, that is, if it maps actions to observations in a way that a computable program can achieve. It is in this sense that, it turns out, AIXI can learn to one-box in *repeated* Opaque Newcomb, and to smoke in *repeated* Smoking Lesion, but may fail all other Newcomblike problems, because we found no way to reduce them in a computable form. However, we still suspect that AIXI can succeed in the repeated settings.
Category: Artificial Intelligence
[11] viXra:2006.0119 [pdf] submitted on 2020-06-14 03:23:52
Authors: Nirmal Tej Kumar
Comments: 11 Pages. Short Communication
[ A General Multi-disciplinary Thermal Mapping + Signal Processing System to Probe (Graphene Quantum Dots + Virus ) based Nano-Bio Sensor for COVID-19 BIO-CHEMICAL INFORMATION PROCESSING w.r.t Theory + Algorithms + Experimentation + Machine Learning as an interesting Suggestion ]
Category: Artificial Intelligence
[10] viXra:2006.0110 [pdf] submitted on 2020-06-12 20:16:52
Authors: Dingbing Li, Yong Deng
Comments: 10 Pages.
Information quality is a concept that can be used to measure the information of probability distribution. Dempster-Shafer evidence theory can describe uncertain information more reasonably than probability theory. Therefore, it is a research hot spot to propose information quality applicable to evidence theory. Recently, Deng proposed the concept of information volume based on Deng entropy. It is worth noting that, compared with the Deng entropy, the information volume of the Deng entropy contains more information. Obviously, it may be more reasonable to use information volume of Deng entropy to represent uncertain information. Therefore, this article proposes a new information quality, which is based on the information volume of Deng entropy. In addition, when the basic probability (BPA) degenerates into a probability distribution, the proposed information quality is consistent with the information quality proposed by Ygare and Petry. Finally, several numerical examples illustrate the effectiveness of this new method.
Category: Artificial Intelligence
[9] viXra:2006.0079 [pdf] replaced on 2021-02-18 16:44:18
Authors: Al-Akhir Nayan, Md. Obaidur Rahman, Ahamad Nokib Mozumder, Mohammod Abul Kashem
Comments: 13 Pages. Published in Multidisciplinary Journal of European University of Bangladesh, 5(1), 2020 [Corrections made by viXra Admin to conform with the guidelines of viXra.org]
Due to the simplicity and capability to alter according to our requirements, the robotics and automation are being used widely in industries. The scheme is aimed to assemble an automatic vehicle by using GPS, which is depended on computer to generate its path coordinate. GPS module is utilized to collect GPS data. The mobile camera encounters the obstacles, machine learning algorithm assists to avoid it and performs real time object detection. The automobile uses the electric motors to spin wheels and has full control of the throttle, steering and breaking. An Arduino device pilots the vehicle following the instructions generated by the computer. Traffic has increased by quite a huge number. Excessive number of vehicles leads to large number of vehicle accidents every day. Driver
issue is also a great difficulty. The ultimate goal of this work is to minimize the possibilities of accidents and to ensure the safety of the passengers. Thus, the vehicles will be useful for blind and handicraft people. But serving this device to the military is the main target so that they can get benefit at the time of danger. The motorized vehicle includes sensors to observe the surroundings. Besides, it can be managed by human beings, manually.
Category: Artificial Intelligence
[8] viXra:2006.0064 [pdf] submitted on 2020-06-08 09:33:39
Authors: Tao Wen, Yong Deng
Comments: 11 Pages.
How to measure the uncertainty in the open world is a popular topic in recent study. Many entropy measures have been proposed to address this problem, but most have limitations. In this series of paper, a method for measuring the information volume of mass function is presented. The fractal property about the maximum information volume is shown in this paper, which indicates the inherent physical meanings of Deng entropy from the perspective of statistics. The results shows the multifractal property of this maximum information volume. Some experiment results are applied to support this perspective.
Category: Artificial Intelligence
[7] viXra:2006.0062 [pdf] submitted on 2020-06-07 12:18:52
Authors: Xiaozhuang Gao, Yong Deng
Comments: 10 Pages.
Negation is an important operation on uncertainty information. Based on the information volume of mass function, a new negation of basic probability assignment is presented. The result show that the negation of mass function will achieve the information volume increasing. The convergence of negation is the situation when the Deng entropy is maximum, namely high order Deng entropy. If mass function is degenerated into probability distribution, the negation of probability distribution will also achieve the maximum information volume, where Shannon entropy is maximum. Another interesting results illustrate the situation in maximum Deng entropy has the same information volume as the whole uncertainty environment.
Category: Artificial Intelligence
[6] viXra:2006.0061 [pdf] submitted on 2020-06-07 13:22:14
Authors: Lipeng Pan, Yong Deng
Comments: 12 Pages.
Dempster-Shafer Evidence theory is an extension of probability theory, which can describe uncertain information more reasonably. Divergence measure is always an important concept in probability theory. Therefore, how to propose a reasonable divergence measurement has always been a research hot spot in evidence theory. Recently, Deng proposed the concept of information volume based on Deng entropy. It is interesting to note that compared with the uncertainty measure of Deng entropy, information volume of Deng entropy contains more information. Obviously, it might be more reasonable to use information volume of Deng entropy to represent uncertainty information. Based on this, in the paper, we combined the characteristics of non-specific measurement of Deng entropy, and propose a new divergence measure. The new divergence measurement not only satisfies the axiom of distance measurement, but also has some advantages that cannot be ignored. In addition, when the basic probability assignment(BPA) degenerates into probability distribution, the measured result of the new divergence measure is the same as that of the traditional Jensen-Shannon divergence. If the mass function is assigned in probability distribution, the proposed divergence is degenerated as Kullback-Leibler divergence. Finally, some numerical examples are illustrated to show the efficiency of the proposed divergence measure of information volume.
Category: Artificial Intelligence
[5] viXra:2006.0037 [pdf] submitted on 2020-06-04 13:35:02
Authors: Jixiang Deng, Yong Deng
Comments: 18 Pages.
In fuzzy set theory, the fuzzy membership function describes the membership degree of certain elements in the universe of discourse. Besides, Deng entropy is a important tool to measure the uncertainty of an uncertain set, and it has been wildly applied in many fields.
In this paper, firstly, we propose a method to measure the uncertainty of a fuzzy MF based on Deng entropy. Next, we define the information volume of the fuzzy MF. By continuously separating the BPA of the element whose cardinal is larger than $1$ until convergence, the information volume of the fuzzy sets can be calculated. When the hesitancy degree of a fuzzy MF is $0$, information volume of the fuzzy membership function is identical to the Shannon entropy. In addition, several examples and figures are expound to illustrated the proposed method and definition.
Category: Artificial Intelligence
[4] viXra:2006.0035 [pdf] submitted on 2020-06-04 15:04:31
Authors: Tao Wen, Yong Deng
Comments: 11 Pages.
How to measure the uncertainty in the open world is a popular topic in recent study. Many entropy measures have been proposed to address this problem, but most have limitations. In this series of paper, a method for measuring the information volume of mass function is presented. The fractal property about the maximum information volume is shown in this paper, which indicates the inherent physical meanings of Deng entropy from the perspective of statistics. The results shows the linear relationship between the maximum information volume and the probability scale. Some experiment results are applied to support this perspective.
Category: Artificial Intelligence
[3] viXra:2006.0028 [pdf] submitted on 2020-06-03 16:12:01
Authors: Yong Deng
Comments: 14 Pages.
Given a probability distribution, its corresponding information volume is Shannon entropy. However, how to determine the information volume of a given mass function is still an open issue. Based on Deng entropy, the information volume of mass function is presented in this paper. Given a mass function, the corresponding information volume is larger than its uncertainty measured by Deng entropy. The so called Deng distribution is defined as the BPA condition of the maximum Deng entropy.
The information volume of Deng distribution is called the maximum information volume, which is lager than the maximum Deng entropy.
In addition, both the total uncertainty case and the Deng distribution have the same information volume, namely, the maximum information volume. Some numerical examples are illustrated to show the efficiency of the proposed information volume of mass function.
Category: Artificial Intelligence
[2] viXra:2006.0025 [pdf] submitted on 2020-06-03 03:35:58
Authors: Nirmal Tej Kumar
Comments: 2 Pages. Short Communication
Probing cryo-Electron Microscopy Images Using Pyramid Representations in the Context of :
[ Image J/ImageJ_Pyramid_Plugin/JikesRVM - Research Virtual Machine(RVM)/JVM - Java Virtual Machine/JI Prolog - Java based Prolog/HPC-High Performance Computing ] for Next Generation Java based[ AI + Image Processing + Informatics ] R&D Test Platforms.
Category: Artificial Intelligence
[1] viXra:2006.0002 [pdf] submitted on 2020-06-01 09:11:02
Authors: Kumar Dron Shrivastav, Neha Taneja, Priyadarshini Arambam, Vandana Bhatia, Shelly Batra, Harpreet Singh, Eyad H. Abed, Priya Ranjan, Rajiv Janardhanan
Comments: 22 Pages. Preprint!
Cervical cancer is a major public health challenge. Further mitigation of cervical cancer can greatly benefit from development of innovative and disruptive technologies for its rapid screening and early detection. The primary objective of this study is to contribute to this aim through large scale screening by development of Artificial Intelligence enabled Intelligent Systems as they can support human cancer experts in making more precise and timely diagnosis. Our current study is focused on development of a robust and interactive algorithm for analysis of colposcope-derived images analysis and a diagnostic tool/scale namely the OM- The Onco-Meter. This tool was trained and tested on 300 In-dian subjects/patients yielding 77% accuracy with a sensitivity of 83.56% and a specicity of 59.25%. OM-The Oncometer is capable of classifying cervigrams into cervical dysplasia, carcinoma in situ (CIS) and invasive cancer(IC). Pro-
gramming language - R has been used to implement and compute earth mover distances (EMD) to characterize different diseases labels associated with cervical cancer, computationally. Deployment of automated tools will facilitate early
diagnosis in a noninvasive manner leading to a timely clinical intervention for
cervical cancer patients upon detection at a Primary Health Care (PHC). The tool developed in this study will aid clinicians to design timely intervention strategies aimed at improving the clinical prognosis of patients.
Category: Artificial Intelligence
| Third-Party Links: |
|