Title: Machine Learning

Abstract:

Voice recognition is a biometric identity authentication technology, also known as speaker recognition. Its theoretical basis is that each voice has a unique feature, which can effectively distinguish the voices of different people. Compared with other biometric technology, voice recognition has no loss and forgetting, does not need to remember, easy to use, and does not involve privacy, therefore, users are easy to accept this technology. Voice recognition can be widely used in security verification, control and other aspects, especially identity recognition in  telecommute application scenarios. In this paper, a GAN-based timbre conversion system is studied and implemented. The forged audio generated by the timbre conversion is used to successfully attack the speaker recognition system. Firstly, according to the GAN, combined with the VCC2016 voice data set, a model which can transform audio files between different timbre is obtained. The model can extract the key features of the audio files from the data set, including Linear Predictive Cepstrum Coefficients (LPCC), MEL Frequency Cepstral Coefficients (MFCC). By means of short-time Fourier transform,  Characteristic parameters get different audio feature, so the GAN makes the network master these laws between the data, thus it may carry on the fitting of audio features to generate the required forged audio files. Through the verification of the existing main speaker recognition algorithms, the method proposed in this paper can effectively attack the existing main speaker recognition technologies, which proves that the security of the current recognition system is defective, and it is urgent to improve its security.

Biography:

Liu peishun, associate professor of ocean university of China, graduated from computer application technology major.His main research interests are artificial intelligence, network and information security. He has participated in many national projects, and published more than 30 papers.

Title: Artificial Intelligence in Healthcare

Abstract:

Introduction: Diagnosis of patients presenting with chronic respiratory symptoms is difficult, because the symptoms of COPD and asthma may be similar, and their diagnostic criteria overlap. However, treatment recommendations for COPD and asthma differ, and inappropriate treatment as a result of misdiagnoses bears the potential to increase the risk of exacerbations, morbidity and mortality, and reduces quality of life. Machine learning (ML) offers an innovative approach of mining large electronic health records data to develop diagnostic algorithms for disease differentiation.

Methods: From a US electronic health records database, covering primary care, specialist care and hospital medical records, cohorts of patients’ ≥35 years who had a specialist diagnosis of asthma, COPD or both (asthma-COPD overlap, ACO) on ≥2 occasions were created. The specialist diagnosis was used as the case label. Over 60 clinical features including spirometry results, blood test results, comorbidities and symptoms were extracted from patients’ electronic health records data within 12-months before and 12-months after patients’ incident diagnosis. Eleven supervised ML methods were investigated to perform disease classification on 85% of the labeled cases, and the remaining 15% were used as a hold out data set for model validation.

Results: A total of 240,378 COPD, 143,748 asthma and 27,437 ACO cases were identified. Extreme Gradient Boosting (XGB) with Bayesian hyper-parameter optimization had the best performance. The XGB model with 12 clinical features including spirometry results, pack-years, body mass index, symptoms, and allergic rhinitis and chronic rhinitis achieved a sensitivity of 0.98, 0.98 and 0.78, and an F1-score (accuracy measure) of 0.98, 0.98 and 0.84, in diagnosing COPD, asthma and ACO, respectively.

Conclusions: Machine learning is a powerful tool to aid physicians in the differential diagnosis of asthma, COPD and ACO. Additional studies are needed to evaluate the model in other settings and countries, and to assess its safety for guiding treatment decisions.

Biography:

Dr. Alan Kaplan is a Honorary Professor of Primary Care in Respiratory Medicine.Family Physician practicing in York Region, Ontario, Canada. Chairperson of the Family Physician Airways Group of Canada. Past- Chairperson of the Respiratory Section of the College of Family Physicians of CanadaSenate member of the International Primary Care Respiratory Group President of the IPCRG 5th biennual world scientific meeting, “Making Every Breath Count” Toronto June 2-5, 2010. Chair of the Council of Organizing Members of the Canadian Network for Respiratory Care. Member of Past Canadian Consensus Guidelines for H pylori, Asthma, COPD and Sinusitis. Representative of the College of Family Physicians of Canada to the SARS clinical working group.

Title: Using Natural Language Processing to Understand Us

Abstract:

The latest popular approach in natural language processing is the so-called word embedding (WE). The word embedding (WE) model is a neural network based on the distributional semantic model. The distributional hypothesis states that semantically similar words tend to have similar contextual distributions. In the WE context, if two words have similar vectors, then they have the same distribution. An application of WE based on periodical is called temporal word embedding or dynamic word embedding (DWE).
We have explored the use of WE and DWE to mine lifestyle, sentiment and evolution of trends and policy. In the work of WE on Malaysian Twitter corpus, we explore the possibility of viewing Malaysia's lifestyle on where they spend most of their time for social meetings and analyse the sentiment of a public figure.

Biography:

Sabrina Tiun is a senior lecturer at the Faculty of Information Science and Technology (FTSM) in Universiti Kebangsaan Malaysia (UKM), Malaysia. Her research interests range from natural language processing, computational linguistic, and speech processing to social science computing. She teaches several courses in natural language processing and data science at both undergraduate and postgraduate levels. She also actively and successfully supervises postgraduate students for master and PhD level in natural language processing and speech processing. Currently, she is the head lab of Asian language processing (ASLAN) lab and the teaching and learning coordinator for the Centre of Artificial Intelligence Technology at FTSM.