1,721,497 research outputs found
Conditional GAN based Collaborative Filtering with Data Augmentation for Cold-Start User
No full textIn this paper, we propose Cold-CFGAN, a collaborative filtering using two Conditional Generative Adversarial Networks (CGANs). In Cold-CFGAN, one CGAN is used for data augmentation of cold-start users, and the other CGAN is used to recommend items using user condition vectors. ColdCFGAN research uses an additional GAN model to generate data for cold-start users to resolve the cold start problem that occurs when implementing CGAN-based collaborative filtering and to further improve the accuracy of the model. To this end, we first identified the performance degradation problem
of cold-start users through a series of preliminary experiments using an existing conditional GAN-based collaborative filtering (CFGAN). Then, we used the user profile and item purchase data to express the number of purchased items per user in the form of a percentile, and identified cold-start users with few purchase items. Using the profile of the identified cold-start user data, we found the data of the Item-Rich user with the most similar profile to the cold-start user based on the cosine similarity, and using the data of the Item-Rich user, we applied partial masking method to create augmented cold-start users. Then we train user augmentation GAN to generate fake Item-Rich user using the augmented cold-start user and corresponding ItemRich user in real data. We use trained generator to generate Item-Rich user corresponding to cold-start user in real dataset. Then, we applied the generated Item-Rich user data to train the conditional GAN-based collaborative filtering and after training,
we performed experiment. Through the experiment, we found improved performance for cold start users compared to the traditional approach, and also improved overall performance
Advanced nonlinear controllers for the chemotherapy of brain tumor
Full text linkBrain tumors represent a significant challenge in the medical field due to their
complex nature and the critical environment in which they develop. These tumors
result from the abnormal proliferation of cells within the brain, leading to the
formation of masses that can be either benign or malignant. The treatment approach
for brain tumors is highly dependent on the type, the location, and the severity of
the tumor. Malignant brain tumors, in particular, present a substantial challenge
due to their aggressive growth patterns and their proximity to essential brain tissues.
Because of the sensitive nature of brain tissue and the intricate location of these
tumors, surgical intervention, which is a common treatment for other types of
tumors, is often not recommended for malignant brain tumors. The reason for
this caution is that malignant tumors are frequently intertwined with critical and
sensitive areas of the brain, making surgical removal risky and potentially leading to
severe neurological damage or loss of critical brain functions.
As a result, alternative treatment strategies, such as chemotherapy, become
crucial for managing malignant brain tumors. Chemotherapy involves the adminis
tration of drugs designed to target and kill rapidly dividing tumor cells; however,
the effectiveness of chemotherapy is heavily reliant on precise dosing and timing.
The goal is to maximize the elimination of tumor cells, while minimizing damage
to healthy brain cells and preserving the patient’s immune function. Striking this
balance is critical because an overdose of chemotherapy drugs can lead to toxicity
and damage to healthy tissues, while an underdose may not be effective in controlling
tumor growth. Therefore, the precise control of chemotherapy dosing is essential to
achieve the best possible treatment outcomes.
This research focuses on developing advanced nonlinear controllers to optimize
chemotherapy dosing. These controllers, including Adaptive Terminal Sliding Mode
Control (AT-SMC), Adaptive Super-Twisting Sliding Mode Control (AS-SMC), Ter
minal Synergetic Control (TSC), Fuzzy Logic Control (FLC), and Barrier Function
Based Sliding Mode Control (BF-SMC), are designed to dynamically adjust the
chemotherapy dosage in response to the tumor’s progression, ensuring that tumor
cells are effectively targeted while healthy cells are preserved.
The controllers are rigorously tested using MATLAB simulations under various
conditions to evaluate their effectiveness in maintaining the balance between eliminat
ing tumor cells and preserving healthy tissue. The stability and convergence of these
systems are verified using Lyapunov theory, which confirmed that the controllers are
capable of achieving the desired outcomes. Among the tested strategies, AT-SMC,
AS-SMC, and BF-SMC showed the most promising results, demonstrating minimal
steady-state error, fast convergence, and efficient drug usage. These findings suggest
that advanced nonlinear control strategies hold significant potential for enhancing
the effectiveness of chemotherapy in treating brain tumors, offering a more targeted
and safe approach to managing this challenging condition
Fuzzy Logic Controller for the Chemotherapy of Brain Tumor
No full textAn advanced Fuzzy Logic Controller (FLC) that considers all the states of the brain tumor system is designed for the chemotherapy treatment. A Mamdani-type FLC is proposed for dynamically controlling the chemotherapy drug for the tumor system; the chemotherapy treatment of brain tumors requires advanced strategies which mainly depend upon the severity of the tumor. In this work, the advanced FLC designed aims both at determining the amount of chemotherapy to eliminate tumor cells, and at preserving the minimum amount of healthy and immune cells. The controller's performance is verified using MATLAB software based on different control parameters, showing its effectiveness in reducing the tumor cells. It has shown favorable results in terms of steady-state error, rate of convergence, and amount of drug consumed
Attention based Remote Photoplethysmography Estimation from Facial Video with Equilibrium in Time-Frequency Supervision
No full textIn pre-clinical health monitoring, estimating physiological signals from video is a low-cost and convenient tool. Remote photoplethysmography (rPPG) involves placing a camera in a remote area to
estimate a person’s heart rate or Blood Volume Pulse (BVP). In this paper, we propose an attention based deep architecture for rPPG estimation that assimilate temporal relationship across a sequence
of frames while focusing on the relevant features and regions by exploiting the inter-pixel relationship of feature maps. Also, we design a dynamic supervision strategy using frequency and time domain
losses to mitigate overfitting for efficient estimation of rPPG signals. The proposed method was evaluated on two publicly available rPPG datasets (UBFC-rPPG and PURE). The findings of this study
demonstrate that promising results can be achieved by enforcing an adequate balance between time-frequency supervision
Deep Representation Learning with Sample Generation and Augmented Attention Module for Imbalanced ECG Classification
No full textDeveloping an efficient heartbeat monitoring system has become a focal point in numerous healthcare applications. Specifically, in the last few years, heartbeat classification for arrhythmia detection has gained considerable interest from researchers. This paper presents a novel deep representation learning method for the efficient detection of arrhythmic beats. To mitigate the issues associated with the imbalanced data distribution, a novel re-sampling strategy is introduced. Unlike the existing oversampling methods, the proposed technique transforms majority-class samples into minority-class samples with a novel translation loss function. This approach assists the model in learning a more generalized representation of crucially important minority class samples. Moreover, by exploiting an auxiliary feature, an augmented attention module is designed that focuses on the most relevant and target-specific information. We adopted an inter-patient classification paradigm to evaluate the proposed method. The experimental results of this study on the MIT-BIH arrhythmia database clearly indicate that the proposed model with augmented attention mechanism and over-sampling strategy significantly learns a balanced deep representation and improves the classification performance of vital heartbeats.
Intumescence behaviour of bottom ash based geopolymer mortar through microwave irradiation – As affected by alkali activation
Full text link
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
- …
