OOOlufade Falade Williams Onifade

Olufade Falade Williams Onifade

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Olufade Falade Williams Onifade has a total of 12 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Lung Cancer Classification from CT Images Using ResNet

    by Olajumoke O. Adekunle, Joseph D. Akinyemi, Khadijat T. Ladoja, Olufade F. W. Onifade

    Lung cancer, a malignancy originating in lung tissues, is commonly diagnosed and classified using medical imaging techniques, particularly computed tomography (CT). Despite the integration of machine learning and deep learning methods, the predictive efficacy of automated systems for lung cancer classification from CT images remains below the desired threshold for clinical adoption. Existing research predominantly focuses on binary classification, distinguishing between malignant and benign lung nodules. In this study, a novel deep learning-based approach is introduced, aimed at an improved multi-class classification, discerning various subtypes of lung cancer from CT images. Leveraging a pre-trained ResNet model, lung tissue images were classified into three distinct classes, two of which denote malignancy and one benign. Employing a dataset comprising 15,000 lung CT images sourced from the LC25000 histopathological images, the ResNet50 model was trained on 10,200 images, validated on 2,550 images, and tested on the remaining 2,250 images. Through the incorporation of custom layers atop the ResNet architecture and meticulous hyperparameter fine-tuning, a remarkable test accuracy of 98.8% was recorded. This represents a notable enhancement over the performance of prior models on the same dataset.

  • Open Access

    Contextual Emotional Transformer-Based Model for Comment Analysis in Mental Health Case Prediction

    by Ayodeji O. J. Ibitoye, Oladosu O. Oladimeji, Olufade F. W. Onifade

    Mental health (MH) assessment and prediction have become critical areas of focus in healthcare, leveraging developments in natural language processing (NLP). Recent advancements in machine learning have facilitated the exploration of predictive models for MH based on user-generated comments that overlooked the integration of emotional attention mechanisms. The methods often struggle with contextual nuances and emotional subtleties, leading to suboptimal predictions. The prevailing challenge lies in accurately understanding the emotional context embedded within textual comments, which is crucial for effective prediction and intervention. In this research, we introduce a novel approach employing contextual emotional transformer-based models (CETM) for comment analysis in MH case prediction. CETM leverages state-of-the-art transformer architectures enhanced with contextual embedding layers and emotional attention mechanisms for MH case prediction. By incorporating contextual information and emotional cues, CETM captures the underlying emotional states and MH indicators expressed in user comments. Through extensive experimentation and evaluation, both Roberta and bidirectional encoder representations from transformers (BERT) models exhibited enhanced accuracy, precision, recall, and F1 scores compared to their counterparts lacking emotional attention. Notably, the Roberta model attained a greater accuracy of 94.5% when matched to BERT’s 87.6% when emotional attention was employed. Hence, by incorporating emotional context into the predictive model, we achieved significant improvements, which offers promising avenues for more precise and personalized MH interventions.


  • Open Access

    Unmasking the Uniqueness: A Glimpse into Age-Invariant Face Recognition of Indigenous African Faces

    by Fakunle Ajewole, Joseph Damilola Akinyemi, Khadijat Tope Ladoja, Olufade Falade Williams Onifade

    The task of recognizing the age-separated faces of an individual, Age-Invariant Face Recognition (AIFR), has received considerable research efforts in Europe, America, and Asia, compared to Africa. Thus, AIFR research efforts have often under-represented/misrepresented the African ethnicity with non-indigenous Africans. This work developed an AIFR system for indigenous African faces to reduce the misrepresentation of African ethnicity in facial image analysis research. We adopted a pre-trained deep learning model (VGGFace) for AIFR on a dataset of 5,000 indigenous African faces (FAGE\_v2) collected for this study. FAGE\_v2 was curated via Internet image searches of 500 individuals evenly distributed across 10 African countries. VGGFace was trained on FAGE\_v2 to obtain the best accuracy of 81.80\%. We also performed experiments on an African-American subset of the CACD dataset and obtained the best accuracy of 91.5\%. The results show a significant difference in the recognition accuracies of indigenous versus non-indigenous Africans.

  • Open Access

    An age-group ranking model for facial age estimation

    by Joseph D. Akinyemi, Olufade F. W. Onifade

    Age prediction has become an important Computer Vision task. Although this task requires the age of an individual to be predicted from a given face, research has shown that it is more intuitive and easier for humans to decide which of two individuals is older than to decide how old an individual is. This work follows this intuition to aid the age prediction of a face by exploiting the age information available from other faces. It goes further to explore the statistical relationships between facial features within age groups to compute age-group ranks for a given face. The resulting age-group rank is low-dimensional and age-discriminatory, thus improving age prediction accuracy when fed into an age predictor. Experiments on publicly available facial ageing datasets (FGnet, PAL, and Adience) reveal the effectiveness of the proposed age-group ranking model when used with traditional Machine learning algorithms as well as Deep Learning algorithms. Cross-dataset validation, a method of training and testing on entirely different datasets, was also employed to further investigate the effectiveness of this method.


  • Predicting Suicide Ideation from Social Media Text Using CNN-BiLSTM

    by Christianah T. Oyewale, Joseph D. Akinyemi, Ayodeji O.J Ibitoye, Olufade F.W Onifade


  • Machine Learning-Based Diabetes Risk Prediction Using Associated Behavioral Features

    by Ayodeji O. J. Ibitoye, Joseph D. Akinyemi, Olufade F. W. Onifade

    Diabetes is a global health concern that affects people of all races. With different uncertainties in human lifestyles, it is difficult to predict diabetes while assuming that the risk patterns are the same for all. The likelihood of diabetes in a patient is mostly predicted using machine learning (ML) models on features explicitly available in datasets, while the intrinsic relationship between features viz-a-viz their potential relevance to the presence of diabetes is oftentimes neglected. In this work, we explored feature importance and correlation to derive the top 15 feature pairs from a dataset of 263,882 samples of anonymized patient information. These top-15 feature pairs were fed into five different ML models (decision tree (DT), neural networks (NN), random forest (RF), support vector machine (SVM) and extreme gradient boosting (XGB)) for predicting the likelihood of diabetes, while also feeding the direct features (without correlated pairing) separately into the same 5[ ]ML models. The models’ performances were evaluated using accuracy, precision, recall and F1-score and NN presented the best performance overall achieving an F1-score of 85% for the correlated feature pairs (CF) and 75% for the direct feature pairs. The results confirm the importance of the correlation/relationship between features in predicting the likelihood of diabetes in patients more accurately.


  • Open Access

    Utilizing RoBERTa Model for Churn Prediction through Clustered Contextual Conversation Opinion Mining

    by Ayodeji O. J. Ibitoye, Olufade F.W. Onifade

    In computational study and automatic recognition of opinions in free texts, certain words in sentences are used to decide its sentiments. While analysing each customer’s opinion per time in churn management will be effective for personalised recommendations. Oftentimes, the opinion is not sufficient for contextualised content mining. While personalised recommendations are time consuming, it also does not provide complete picture of an overall sentiment in the business community of customers. To help businesses identify widespread issues affecting a large segment of their customers towards engendering patterns and trends of different customer churn behaviour, here, we developed a clustered contextualised conversation as opinions set for integration with Roberta Model. The developed churn behavioural opinion clusters disambiguated short messages while charactering contents collectively based on context beyond keyword-based sentiment matching for effective mining. Based on the predicted opinion threshold, customer churn category for group-based personalised decision support was generated, with matching concepts. The baseline RoBERTa model on the contextually clustered opinions, trained with a batch size of 16, a learning rate of 2e-5, over 8 epochs, using a maximum sequence length of 128 and standard hyperparameters, achieved an accuracy of 92%, Precision of 88%, Recall of 86% and F1 score of 84% over a test set of 30%.


  • A Deep Multi-Layer Perceptron Model for Automatic Colourisation of Digital Grayscale Images

    by Olawande M. Shokunbi, Joseph Damilola Akinyemi, Olufade Falade Williams Onifade

  • Open Access

    An Individualized Face Pairing Model for AgeInvariant Face Recognition

    by Joseph Damilola Akinyemi, Olufade Falade Williams Onifade

    Among other factors affecting face recognition and verification, the aging of individuals is a particularly challenging one. Unlike other factors such as pose, expression, and illumination, aging is uncontrollable, personalized, and takes place throughout human life. Thus, while the effects of factors such as head pose, illumination, and facial expression on face recognition can be minimized by using images from controlled environments, the effect of aging cannot be so controlled. This work exploits the personalized nature of aging to reduce the effect of aging on face recognition so that an individual can be correctly recognized across his/her different age-separated face images. To achieve this, an individualized face pairing method was developed in this work to pair faces against entire sets of faces grouped by individuals then, similarity score vectors are obtained for both matching and non-matching image-individual pairs, and the vectors are then used for age-invariant face recognition. This model has the advantage of being able to capture all possible face matchings (intra-class and inter-class) within a face dataset without having to compute all possible image-to-image pairs. This reduces the computational demand of the model without compromising the impact of the ageing factor on the identity of the human face. The developed model was evaluated on the publicly available FG-NET dataset, two subsets of the CACD dataset, and a locally obtained FAGE dataset using leave-one-person (LOPO) cross-validation. The model achieved recognition accuracies of 97.01%, 99.89%, 99.92%, and 99.53% respectively. The developed model can be used to improve face recognition models by making them robust to agevariations in individuals in the dataset.


  • Open Access

    Improved customer churn prediction model using word order contextualized semantics on customers’ social opinion

    by Ayodeji O. J. Ibitoye, Olufade F. W. Onifade

    Through the hype in digital marketing and the continuous increase in volume and velocity of opinions about an organization’s brands, churn prediction now requires advanced analytics in opinion mining for effective customer behavioral management beyond keywords sentiment analysis (SA). Earlier, by analyzing customers’ opinions using SA models, the extracted positive-negative polarity is used to classify customers as churners or non-churner. In those methods, the impact of word order, context, and the inherent semantics of the clustered opinion set were oftentimes overlooked. However, with the consistent creation of new words with new meanings mapped to existing words on the web, the research extended the fuzzy support vector model (FSVM) to show that the dependency distance between the headword, its dependent, and tail word can be weighted by using information content derived from a corpus to generate four-classed social opinion categories as a strongly positive, positive, negative, and strong negative. These opinion classes formed the basis for the churn category as a premium customer, Inertia customer potential churner, and churner in customer behavioral management. In performance evaluation, aside from engendering quadrupled churn class against the existing churn binary pattern, better accuracy, precision, and recall values were obtained when compared with existing SA works in support vector machine and fuzzy support vector machine (FSVM), respectively.