Search
Search Results
-
LSI with Support Vector Machine for Text Categorization – a practical example with Python
18-29Views:558Artificial intelligence is becoming a powerful tool of modernity science, there is even a science consensus about how our society is turning to a data-driven society. Machine learning is a branch of Artificial intelligence that has the ability to learn from data and understand its behavers. Python programming language aiming the challenges of this new era is becoming one of the most popular languages for general programming and scientific computing. Keeping all this new era circumstances in mind, this article has as a goal to show one example of how to use one supervised machine learning method, Support Vector Machine, and to predict movie’s genre according to its description using the programming language of the moment, python. Firstly, Omdb official API was used to gather data about movies, then tuned Support Vector Machine model for Latent semantic indexing capable of predicting movies genres according to its plot was coded. The performance of the model occurred to be satisfactory considering the small dataset used and the occurrence of movies with hybrid genres. Testing the model with larger dataset and using multi-label classification models were purposed to improve the model.
-
Innovative Strategies and Student Academic Performance: Machine Learning Insights on International Students in Chinese Universities
37-60Views:211The higher education sector in China has faced unprecedented challenges recently due to the global COVID-19 pandemic. The influx of international students, a vital component of the nation's academic landscape, presented distinct challenges, including maintaining academic achievements through various online platforms, which necessitated innovative strategies to ensure that their educational pursuits remained rewarding despite these challenges. This study aims to explore the innovative strategies adopted by Chinese higher education institutions in response to the COVID-19 pandemic and examine their impact on the academic achievements of international students. This study employs a comprehensive approach that incorporates questionnaire surveys and dominant Machine Learning Algorithms, such as Multiple Linear Regression (MLR), Decision Tree Model (DTM), Support Vector Regression Model (SVRM), and K-nearest neighbors (KNN). By employing rigorous data-gathering approaches, our study aimed to address a set of particular questions: How did these innovative strategies improve students' academic performance in the face of environmental emergencies? To what extent did international students benefit from these adaptations? Through investigation of these concerns, our research provides insight into the effectiveness of these strategies and their possible significance for future educational methodologies. Innovative strategies positively correlated with student academic performance during the COVID-19 pandemic in Chinese higher Education. This research highlights how overcoming these challenges can have broader implications for shaping resilient global education systems in future crises. The study accurately predicted academic performance, highlighting the importance of innovative teaching approaches in the context of the COVID-19 pandemic. This study might influence educational policies and practices. Educational institutions can make informed decisions about emergency preparedness and development by assessing results using a creative approach. Our findings bring depth to the global conversation on higher Education under challenging circumstances, showing how Innovation might alleviate the adverse impacts on international students' learning experiences.
-
Battery Measurement Methods and Artificial Intelligence Applied in Energy Management Systems
428-436Views:244Diagnostics of batteries using advanced methods have gained remarkable roles in the past few years. This study focuses on the type of measurements, tests and methods to reveal and classify them. During manufacturing and operation several faults could emerge in batteries including non-optimal operation conditions, operators without experience, and finally, random changes in batteries under physical and nonphysical conditions. Improper handling of batteries and battery cells man cause operation failures or, in the worst case, accidents. To reveal these problems several methods are applied in industry and in scientific laboratories. For a comprehensive analysis of battery management, artificial intelligence and Industry 4.0 methods can be used very effectively. Big Data analysis in its standard form is not a new achievement, but other mathematical tools could be applied to control monitoring such as Fuzzy Logic or Support Vector Machine (SVM). They are efficient tools to analyse the deviation of batteries condition because it can detect sudden changes, parameter deviations and anomalies, and the user’s behaviour and habits. This article gives a description about the most important battery testing methods and the connection between Big Data and Operation Management with Artificial Intelligent (AI) methods.
