Sotiris K. Tasoulis

Sotiris K. Tasoulis

Auto-generated profile, based on publicly available information

Sotiris K. Tasoulis

University of Thessaly

Research Focus & Metrics

Academic Career Length

9Years (2018-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
28 October 2020 in Information
Verified
Journal Article
Open access
PDF file
Aristidis VrahatisAristidis Vrahatis
Aristidis Vrahatis
STSotiris K. Tasoulis
Sotiris K. Tasoulis
Spiros V. GeorgakopoulosSpiros V. Georgakopoulos
Spiros V. Georgakopoulos
Vassilis PlagianakosVassilis Plagianakos
Vassilis Plagianakos
Nowadays, biomedical data are generated exponentially, creating datasets for analysis with ultra-high dimensionality and complexity. An indicative example is emerging single-cell RNA-sequencing (scRNA-seq) technology, which isolates and measures individual cells. The analysis of scRNA-seq data consists of a major challenge because of its ultra-high dimensionality and complexity. Towards this direction, we study the generalization of the MRPV, a recently published ensemble classification algorithm, which combines multiple ultra-low dimensional random projected spaces with a voting scheme, while exposing its ability to enhance the performance of base classifiers. We empirically showed that we can design a reliable ensemble classification technique using random projected subspaces in an extremely small fixed number of dimensions, without following the restrictions of the classical random projection method. Therefore, the MPRV acquires the ability to efficiently and rapidly perform classification tasks even for data with extremely high dimensionality. Furthermore, through the experimental analysis in six scRNA-seq data, we provided evidence that the most critical advantage of MRPV is the dramatic reduction in data dimensionality that allows for the utilization of computational demanding classifiers that are considered as non-practical in real-life applications. The scalability, the simplicity, and the capabilities of our proposed framework render it as a tool-guide for single-cell RNA-seq data which are characterized by ultra-high dimensionality. MRPV is available on GitHub in MATLAB implementation.


27 February 2018 in arXiv
Verified
Preprint
Open access
PDF file
Spiros V. GeorgakopoulosSpiros V. Georgakopoulos
Spiros V. Georgakopoulos
STSotiris K. Tasoulis
Sotiris K. Tasoulis
Aristidis VrahatisAristidis Vrahatis
Aristidis Vrahatis
Vassilis PlagianakosVassilis Plagianakos
Vassilis Plagianakos
Flood of information is produced in a daily basis through the global Internet usage arising from the on-line interactive communications among users. While this situation contributes significantly to the quality of human life, unfortunately it involves enormous dangers, since on-line texts with high toxicity can cause personal attacks, on-line harassment and bullying behaviors. This has triggered both industrial and research community in the last few years while there are several tries to identify an efficient model for on-line toxic comment prediction. However, these steps are still in their infancy and new approaches and frameworks are required. On parallel, the data explosion that appears constantly, makes the construction of new machine learning computational tools for managing this information, an imperative need. Thankfully advances in hardware, cloud computing and big data management allow the development of Deep Learning approaches appearing very promising performance so far. For text classification in particular the use of Convolutional Neural Networks (CNN) have recently been proposed approaching text analytics in a modern manner emphasizing in the structure of words in a document. In this work, we employ this approach to discover toxic comments in a large pool of documents provided by a current Kaggle's competition regarding Wikipedia's talk page edits. To justify this decision we choose to compare CNNs against the traditional bag-of-words approach for text analysis combined with a selection of algorithms proven to be very effective in text classification. The reported results provide enough evidence that CNN enhance toxic comment classification reinforcing research interest towards this direction.