SOTONWA KEHINDE ADEBOLA

Computer Science

At the Computer Science department office

Appointment on Visitation important

Topic: Development Of Improved Schema Entropy And Interface Complexity Metrics

Description:

Introduction: Extensible Markup Language is a mark-up language that defines a set of rules for encoding documents in a format that is both human-readable and machine-readable for managing information contained in the schema documents and exchanging wide variety of elements class of data to reflect understandability and reusability.

Aim: to develop improved schema entropy and interface complexity metrics.

Objectives: formulate schema entropy and interface complexity metrics using RNG, verify the metrics using Weyuker's properties and compare the performance of the metrics with existing metrics to prove its robustness.

Methodology: The developed schema entropy and interface complexity metrics acquired 40 schema documents from WSDL and implemented in RNG. The metrics are based on the concept from DTD metrics to measure complexity of schema on similarly structured elements, distinct structured elements and their occurrences using the Number of Attributes NOA , Number of Equivalence Class NOC , Frequency Occurrence of the class FOC and Number of Elements NOE.

Expected result: A higher value of ISEM and IICM tend to a degree of high flexibility and reusability quality. The empirical validation of the metrics is applied on 40 WSDL schema files. A comparison with similar measures is also performed. The proposed ISEM and IICM metrics validated practically and theoretically. The comparative study proves the robustness of the metrics and performed better in terms of reusability and reducing lengthy code. The statistical analysis of the study showed a significant and linear relationship, and high degree of correlation.

Contribution to Knowledge: this research contributes to understandability of schema documents in all schema languages for easy access of XML documents, quality and maintenance in distributed application and providing feedback in software construction project design.

#	Certificate	School	Year
1.	Ph.D (Computer Science)	Computer Engineering/ Ladoke Akintola University of Technology, Ogbomoso, Oyo State	2020

Development of deep learning model in breast cancer using digital pathology

There s a surge of interest in using deep learning, a type of artificial intelligence, to analyze digital pathology images for breast cancer detection and diagnosis. This fascination stems from the potential of deep learning models to: -Improve accuracy and efficiency: Deep learning can automatically identify patterns in complex images that may be missed by the human eye, potentially leading to earlier and more precise diagnoses. -Assist pathologists: Models can act as a second opinion for pathologists, highlighting suspicious areas and reducing workload. Develop personalized medicine: By analyzing tumor characteristics, deep learning could aid in predicting a cancer s aggressiveness and guide personalized treatment plans. Researchers are actively developing deep learning models for various breast cancer tasks. Some areas of focus include: -Distinguishing between benign and malignant tumors. -Assessing the severity of cancer based on factors like cell growth. -Identifying specific breast cancer subtypes for targeted therapies. Overall, deep learning in digital pathology holds promise for revolutionizing breast cancer diagnosis and treatment. However, challenges like data privacy and ensuring model fairness require ongoing research.

SOTONWA KEHINDE is a Lecturer I at the Department of Computer Science

SOTONWA has a Ph.D in Computer Science from Computer Engineering/ Ladoke Akintola University of Technology, Ogbomoso, Oyo State