1. Big-Data Analytics
1.1. Multi-modal Emotion Quantification and Analysis
Interpretable Multimodal Emotion Quantification System (I-MEQ) : A Dynamic Statistical Approach to Emotion Monitoring in Video Content
1.2. Natural Language Processing
Natural Language Processing (NLP) is an increasingly important field in artificial intelligence (AI). NLP helps machines understand human language and has been used in a variety of applications such as analyzing product and service user reviews. This technology is used to understand and process human language into data to understand the meaning, intent, and sentiment.
Opinion mining or commonly known as sentiment analysis aims to extract the sentiment orientation of textual data. It derives whether the text expresses a positive, negative, or neutral sentiment orientation. It can be divided into three levels: document level, sentence level, and fine-grained level. In particular, fine-grained opinion mining determines the sentiment on aspects that are explicitly and implicitly expressed.
Our research lab aims to expand the NLP study for low-resource languages (e.g. Korean, Filipino, etc.) in developing fine-grained opinion mining for business applications. We aim to contribute to NLP research by expanding research lexical semantics and handling knowledge based on I.Q. (word definitions), E.Q. (opinion mining), and C.Q. (contextual and cultural nuances).
1.2. Process Analytics - Process Mining
Process analytics is an analytic methodology that is used for understanding, managing, and improving business processes by defining and explaining each factor and the relationship between the factors. When analyzing past data by the method, it can get some information that can know only from the field; the information cannot recognize from the related documents. According to analytics based on the field’s situations, it can solve problems practically.
We study about 1. Process visualization, diversification of the analytic viewpoint, and methods of drawing processes using data, 2. Conformance analytics of the process model and method of processing the missing anomaly data using the results, 3. Process optimization and simulation according to the drawn process and the analytic results. According to the topics, the lab discovers and deals with the processes, and researches the application plans used the drawing results.
2. Operational Manageability - reliability and resilience analysis
Operational Resilience: is usually defined as the ability of an organization to adapt rapidly to changing environments or unpredictable processes. It is an organization's ability to detect, prevent, respond to, recover and learn from operational disruptions that may impact the delivery of important business and economic functions or underlying business services. The key components of operational resilience - which include defining and understanding important business services and impact tolerance, as well as completing end-to-end mapping, scenario testing, and regular self-assessments - are essential guideposts on the road to resiliency.
3. Productivity evaluation & Benchmarking with DEA
Data envelopment analysis (DEA) should be viewed as a method (or tool) for data-oriented analytics. DEA is a data-driven tool for performance evaluation and benchmarking. While computational algorithms have been developed to deal with a large volume of data (decision-making units, inputs, and outputs) under the conventional DEA, valuable information hidden in big data that are represented by network structures should be extracted by DEA. These network structures encompass a broader range of metrics that cannot be modeled by the conventional DEA. It is shown that network DEA is different from the standard DEA, although it bears the name of DEA and has some similarities with the conventional DEA. Network DEA is a big Data Enabled Analysis (big DEA) of data when multiple (performance) metrics or attributes are linked through network structures. These network structures are too large or complex to be dealt with by the conventional DEA. Unlike the conventional DEA that is solved via linear programming, general network DEA corresponds to nonconvex optimization problems. This represents opportunities for developing techniques for solving non-linear network DEA models.