School of Business and Management Department of Information Systems, Business Statistics and Operations Management 185 Digital Leadership and Digital Transformation Supervisor: LEE, Dongwon / ISOM Student: YANG, Tiffany / ISD Course: UROP1100, Fall This paper examines the similarities and differences of the characteristics between digital leaders from leading companies in the APAC region and the EMEA/NA region. Through examining a total of 24 digital leaders, this study discovers that there are many shared commonalities between these individuals, particularly in their ability to lead their team with a transformative vision and facilitating cross competencies with others both inside and outside their ecosystem. While there exists some slight nuances and differences in the way they lead digital transformation in their own companies, it seems that the contrast between leaders in APAC and EMEA/NA are less pronounced if narrowed to the leading firms in the digital transformation space today. Deep Learning NLP in finance Supervisor: YANG, Yi / ISOM Student: SO, Tung Tung / RMBI Course: UROP1100, Summer Analyzing the sentiment of the entities within financial news articles can be valuable. It allows the trading algorithm to react quickly to the market and may generate potential profit. This kind of entity-level sentiment analysis is an extension of Named Entity Recognition (NER) in natural language processing (NLP) tasks. This report will analyze the 4000 examples annotated by the previous students. Discussing how we do data preprocessing and cleaning. We are providing a brief overview of BERT, how to tokenize the data and fine-tune the model. Finally, we will also evaluate the model's performance based on some traditional metrics. Deep Learning NLP in finance Supervisor: YANG, Yi / ISOM Student: WANG, Xiaopeng / DSCT Course: UROP2100, Fall Existing NLP systems rarely separate numbers and words in the text when embedding them. This obviously contrasts with the consensus in neuroscience that the human’s brains treat numbers differently from words. For instance, most NLP models are not specially designed for processing numbers, most of which either discard numbers directly or replace them with UNK tokens. However, given the ubiquity of numbers especially in the financial area, it would be beneficial for us to enable NLP models to understand numbers more effectively. In this project, I have tried several methods to improve numeracy, which means the ability to understand and work with numbers in either digit or word form (Spithourakis and Riedel,2018), in the real-world application.
RkJQdWJsaXNoZXIy NDk5Njg=