p Dimitris Mallis personal porfolio

This is a personal website for Dimitrios Mallis, Electrical Engineer and Computer Engineer

Short Bio

Dimitrios Stefanos Mallis is an graduate Electrical and Computer Engineer. He received his B.S at Democritus University of Thrace, which he attended from 2010 to 2015. In this time was involved with several academic projects with most notable, his work regarding Sentiment Analysis of Greek Tweets which has been presented to the 19th Panhellenic Conference on Informatics and to the 8th Electrical and Computer Engineering Student Conference of 2015. Also a synopsis of the novelties of his diploma thesis entitled "Separation of Audio Sources in real reverberant rooms", was presented to the 12th AIAI in Thessaloniki Greece and is currently also accepted for publication in a following issue of Springer's “Evolving Systems”.

He was a summer intern at ATEKE Construction S.A from 2010 to 2015 and conclude his military service in 2016. After his graduation he has focused on software engineering for the enterprise. Currently he is a Junior Software Engineer in Singular Logic, the biggest Software Vendor and one of the largest Integrated IT Solutions Groups in Greece, developing mostly server side application and APIs with .NET Technologies.

He is fascinated about the principles of Object Oriented Design and Software Architecture and constantly works upon improving his skills and learning more about the ever changing world of software development.He lives in Athens Greece and enjoys programming, motorcycling and Crossfit. Dimitris Mallis resume can be found here

Publications

Working Experience

  • Junior Software Developer at Singular Logic.

    from 2016 to present

    Designing custom solution and enterprise application for clients. His main focus is in server side applications, middlewares and APIs, using .NET and C#.

  • Summer Intern at ATEKE Construction S.A.

    from 2010 to 2015.

    During his intern sip, he has been in contact with multiple aspects of working in the Construction Industry and also dealt with network maintenance, technical support and various programming tasks.

Notable Academin Projects

  • Sentiment Analysis of Greek Tweets and Hashtags using Sentiment Lexica Pending Publication for the Transactions on Intelligent Systems and Technology. Abstract: "The rapid growth of social media has rendered opinion and sentiment mining an important area of research with a wide range of applications. We focus on the Greek language and the microblogging platform Twitter, investigating methods for extracting sentiment of individual tweets as well as population sentiment for different subjects (hashtags). We propose and investigate the use of sentiment lexicon-based methods as a mean of extracting sentiment information from social media. We compare several approaches for measuring the intensity of six sentiments: anger, disgust, fear, happiness, sadness, and surprise. To evaluate the effectiveness of our methods, we develop a benchmark dataset of tweets, manually rated by two humans. Our automated sentiment results seem promising and correlate to real user sentiment. At this point we make some interesting observations about the lexicon-based approach which lead to the need for a new, better, sentiment lexicon. We then develop a training set of tweets, manually annotated by two independent raters. We ‘transfer’ the sentiment information of these annotated tweets, in a meaningful way, to the set of words that appear in them. In this way, we develop a new sentiment lexicon for use in web applications. We then assess the performance of our methods with the new lexicon and find improved results. Finally, we examine the variation of sentiment intensity over time for selected hashtags, and associate it with real-world events"
  • Skin Detector Matlab App - June 2014 The purpose of the application is to detect Skin Color under different Color Spaces. To manage that , we train an one class classifier from a collection of Skin Patches. The user has to load a folder containing jpg images of skin color before the training of the classifier. For the training of the classifier we take the values of the skin pixels, and for each dimension of the color space, we create a Gaussian Distribution which we use in the next step for the classification. After that we insert an Image and for every pixel of the image and the classifier decides if it is a skin or a non skin pixel. The processed Image is shown into the screen. The result is strongly dependent on the training set and the color model in which the whole process is taking place. The User can use different Color spaces and Sigma Values for his experiments in order to take the best results.
  • Online Forum Database Online Forum Database hosted in a LAMP (Linux with Apache,MySQL,PHP) server. Web application designed in traditional HTML and CSS, PHP for server interaction as well as some Javascript. The main purpose of this project was the familiarization of the students with traditional Data Bases and because of that the web designing part may appear a little poorly implemented.
  • Travelling Salesman Problem with Monte Carlo Tree Search. A C# solution to the TSP problem using Monte Carlo tree Search. The travelling Salesman project description is: Given a list of cities and the distances between each pair of them , what is the shortest possible route that goes through each city exactly once and returns to the first city ? The TSP problem is one of the most important problems of theoretical computer Science and also NP-hard. In this project we solve this problem using Monte Carlo Tree Search. Monte Carlo Tree Search (MCTS) is a method for making optimal decisions in artificial intelligence (AI) problems, typically move planning in combinatorial games. It combines the generality of random simulation with the precision of tree search. In this project we provide a C# solution to the travelling Salesman Problem Using the MCTS accompanied with a UI interface which is helpful for the better understanding of the aspects and functions of the algorithm.
  • Diploma Thesis "Source Separation in Real Room Enviroments" . This diploma thesis, tackles the problem of audio source separation in real room enviroments. ICA framework is used to deal with the convolutive problem at the frequency domain. We examine a new ICA implemetation named RobustICA, that has not been used for audio separation. For the permutation ambiguity problem a new method that is proposed, that is more efficient, which we name Reduced Likelihood Ratio Jump as well as Beamforming teckniques. Finally experiments are performed to asses the affectiveness of our methods from an audio recording dataset that we created and give as a resource from this webpage.

Resources

  • Benchmark Dataset of Greek Tweets. To implement and examine our methods we created a benchmark dataset with Greek tweets, along with a set of manually rated tweets for their sentiment intensity. We make this benchmark Greek dataset publicly available as a valuable resource for future research.
  • Dataset of Mixed Audio Sources. This Audio dataset was created for the needs of the diploma thesis entitles Source ""Separation in Real Room Enviroments"", and contains mixed speech and music sources for the purpose of audio separation. There are recordings of 2, 3 and 4 audio sources, spread in a real room in various positions. This dataset can be used freely in further research for evaluation of audio source separation methods.
  • The new Greek Sentiment Lexicon. This a stemmed (Greek stemmer of Ntais 2006) lexicon contains sentimental rating for greek entries, that are produced from a semi automatic method describded in our paper "Sentiment Analysis of Greek Tweets and Hashtags using Sentiment Lexica" and are speciallized for sentiment analysis of microblogging data.

Interests - Skills

    Topics of interest that he has encounter in his academic path and related skills that has been aquared.

  • .NET : C#, ASP.NET Web Api, LINQ, Ninject, TFS, ADO.NET, Async Programming, CLR.


  • Design And Architecture: Object Oriented Design, SOLID Principles, Enterprise Application Architecture, Design Pattern, Algorithms and Datastructures, TSQL and Microsoft Server, API design, Refactoring


  • General Programming Skills: Json and Json.Net, XML, Http, Git, Javascript, HTML5, AJAX, networking


  • General Academic Skills : Signal processing, Pattern Recognition, Information Retreival, Sentiment Analysis, Hadoop, Map Reduce, Latex, Arithmetic - Numerical Analysis, Probability theory.


  • Strengths: Attitude for constant learning and improvement, Analytical thinking and ability to break Big Problems into Small Modules, Effort Towards Deep Understanding of Studying Subject, Always Seek to produce Software of High Quality through constant Refactoring, Able to Do Independent Work, Research Experience, Effort towards increasing Productivity, Able Communicator

Attended Conferences

  • ECESCON 8, Electrical and Computer Engineering Student Conference 8 Patra, April 2015.
  • Plan B: Hack the system, International Workshop on Internet Security, eestec LC Xanthi - Democritus Univercity of Thrace, March 2015.
  • ECESCON 7 , Dismantling the Problem: The Engineers of the Modern Days, Electrical and Computer Engineering Student Conference 7 Thessaloniki, Aprl 2014.
  • ECESCON 6, Ιnnovation, the answer of the Electrical Engineerer to the modern dilemmas, Electrical and Computer Engineering Student Conference 6 Athens, April 2013.
  • ECESCON 5 ,Electrical and Computer Engineering Student Conference 5 Xanthi,April 2012.
  • Sound and Digital Sound Processing, IEEE Student Branch of Thrace, May 2012.
  • Digital Image and Image Processing, IEEE Student Branch of Thrace, May 2011.
  • PCI 2015. 19th Panhellenic Conference on Informatics. 1-3 October 2015, Athens, Greece.

Hobbies

  • Photography, photoshop
  • Books, reading, audiobooks
  • cinema
  • cooking
  • Guitar
  • Gaming
  • Basketball, KickBoxing
  • Crossfit

<