[서평]「The Unified Modeling Language User Guide」

Give us 5 minutes and we will show you the best book to read today. This is it, the handbook of data mining and knowledge discovery that will be your best choice for better reading book. Your five times will not spend wasted by reading this website. You can take the book as a source to make better concept. Referring the books that can be situated with your needs is sometime difficult. But here, this is so easy. You can find the best thing of book that you can read.

Handbook of Data Mining and Knowledge Discovery

Understanding the sentiment of a comment from a video or an image is an essential task in many applications. Sentiment analysis of a text can be useful for various decision-making processes. One such application is to analyse the popular sentiments of videos on social media based on viewer comments. However, comments from social media do not follow strict rules of grammar, and they contain mixing of more than one language, often written in non-native scripts. Non-availability of annotated code-mixed data for a low-resourced language like Tamil also adds difficulty to this problem. To overcome this, we created a gold standard Tamil-English code-switched, sentiment-annotated corpus containing 15,744 comment posts from YouTube. In this paper, we describe the process of creating the corpus and assigning polarities. We present inter-annotator agreement and show the results of sentiment analysis trained on this corpus as a benchmark.

Corpus creation for sentiment analysis in code-mixed Tamil-English text

Advances in information and signal processing technologies have a significant impact on autonomous driving (AD), improving driving safety while minimizing the efforts of human drivers with the help of advanced artificial intelligence (AI) techniques. Recently, deep learning (DL) approaches have solved several real-world problems of complex nature. However, their strengths in terms of control processes for AD have not been deeply investigated and highlighted yet. This survey highlights the power of DL architectures in terms of reliability and efficient real-time performance and overviews state-of-the-art strategies for safe AD, with their major achievements and limitations. Furthermore, it covers major embodiments of DL along the AD pipeline including measurement, analysis, and execution, with a focus on road, lane, vehicle, pedestrian, drowsiness detection, collision avoidance, and traffic sign detection through sensing and vision-based DL methods. In addition, we discuss on the performance of several reviewed methods by using different evaluation metrics, with critics on their pros and cons. Finally, this survey highlights the current issues of safe DL-based AD with a prospect of recommendations for future research, rounding up a reference material for newcomers and researchers willing to join this vibrant area of Intelligent Transportation Systems.

/pdf/deep-learning-for-safe-autonomous-driving-current-challenges-3u5n6rw5eh.pdf

Deep Learning for Safe Autonomous Driving: Current Challenges and Future Directions

In this paper, we contribute to a recent and successful modelization of uncertainty, which the practitioner often encounters in the formulation of multicriteria group decision making problems. To be precise, in order to approach the uncertainty issue we introduce a novel type of soft rough covering by means of soft neighborhoods, and then we use it to improve decision making in a multicriteria group environment. Our research method is as follows. Firstly we introduce the soft covering upper and lower approximation operators of soft rough coverings. Then its relationships with well-established types of soft rough coverings are analyzed. Secondly, we define and investigate the measure degree of our novel soft rough covering. With this tool we produce a new class of soft rough sets. Finally, we propose an application of such soft rough covering model to multicriteria group decision making by means of an algorithmic solution. A fully developed example supports the implementability of this decision making method.

A novel type of soft rough covering and its application to multicriteria group decision making

Recently, risk based prediction models in medical diagnostic systems gain wider significance in deciding most appropriate diagnostic treatments and for clinical usage. Prostate cancer is a disease which is difficult to diagnose and there are number of failure cases reported. Therefore, an effective and aggressive selection of multiple factors influence on the disease is required. In this paper, an adaptive soft set based diagnostic risk prediction system is presented with the implementation on prostate cancer. The system receives input parameters related to the disease and gives out the risk percentage of the patient. Soft sets are generated with the input parameters by fuzzification followed by rule generation. The risk percentage of the rules are individually calculated for Precision, Recall and F-Measure, that conclude on the best risk percentage based on the maximum area under the curve (AUC) in each case. This ensures to select the most influential risk parameters in treating the disease. Specificity and sensitivity of the test system yield 75.00% and 45.45% respectively.

An Adaptive Soft Set Based Diagnostic Risk Prediction System

Wireless data communication along with data classification techniques has got wider acceptance in various marine wireless applications. This paper exploits the power of machine learning algorithm to classify wireless communication dataset for effective decision making in marine sector. Fishing is among the most risky of professions in the world because once out on the sea, the fishermen are subject to various oceanographic conditions. The unreliable communication between the fishing fleets and to the shore is a serious problem when they face emergency situations like bad weather, border attacks, natural calamities etc. This paper is intended to develop an algorithm to determine the most influential parameters by considering signal strength, wind speed etc. which helps to track, classify and disseminate information to the fishing fleets while they are in deep sea. A decision tree based classification is proposed to find the best node based on the signal strength and the environmental conditions and the scenario has been simulated using NS2 platform. An ensemble based learning algorithm with bagging and adaptive boosting in C4.5 is also employed for improving the performance. The performance comparison has been done and the result shows that the boosted decision tree algorithm has got highest classification accuracy of 95.73%.

Decision tree based data classification for marine wireless communication

Morph Analyser for Malayalam Language: A suffix stripping approach

Cyber security is a most important trepidation in the widespread adoption of internet technologies in the everyday activities of human being. Even though more sophisticated technologies emerged on the Internet, but different kinds of attacks and threats are also increasing day by day. Cyber attacks causes loss of customer confidence in adopting internet based applications. Phishing attack is one of the common vulnerabilities in the cyber space. Most of the anti-phishing solutions proposed so are focused only on a single issue and needs improvement. For malicious web page detection and prevention, an intelligent multi agent solution is proposed in this paper with the help of machine learning methods. The proposed approach detects both phishing sites and websites with malicious content. This multi-agent system contains four autonomous intelligent agents, which communicate with each other using the Extensible Messaging and Presence Protocol (XMPP) for decision-making. The first is a monitoring agent, second and third is for decision-making (using the machine-learning classifiers) and the fourth is for action-performing. The first agent is responsible for extracting URLs. It passes the extracted URLs to the second agent for feature extraction and classification. If any phishing is detected, the second agent communicates with the fourth agent and the site is blocked. Otherwise, the second agent communicates with the third agent for malicious script detection. If any malicious script is detected then the fourth agent blocks the entire web page. We have tested the performance and accuracy of the proposed method and obtained results ensures its efficiency.

An Intelligent System for Phishing Attack Detection and Prevention

Speech being a natural mode of communication for humans is still a challenging problem for Speech Recognition (SR) using a computer. Recently the advancement of technology and deep learning models made a tremendous advancement in Automatic Speech Recognition (ASR) and applied in different regional languages also. This study explored Automatic Speech Recognition using KALDI for continuous Malayalam short stories with different algorithms. Malayalam is the native language of Kerala state in Indian sub-continent. Very less amount of transcribed data is available for Malayalam, so it is recognized as a low resource language. The database consists of twenty Malayalam short stories with transcription. For building such an ASR system the Mel Frequency Cepstral Coefficient (MFCC) features are excerpted with its transformation like Linear Discriminant Analysis (LDA), Maximum Likelihood Linear Transform (MLLT) are extracted from the Malayalam speech database. SGMM + DNN system is modeled and trained for speech utterances.The model’s output is measured and compared to the Word Error Rate (WER). A substantial decrement in WER is reported for the DNN model.

Elizabeth Sherly

Papers

An Adaptive Soft Set Based Diagnostic Risk Prediction System

Decision tree based data classification for marine wireless communication

Morph Analyser for Malayalam Language: A suffix stripping approach

An Intelligent System for Phishing Attack Detection and Prevention

An ASR System for Malayalam Short Stories using Deep Neural Network in KALDI