Dennis Doubovski ★

Dennis Doubovski ★

Netherlands
1K followers 500+ connections

About

My passion is getting from problem to market using computer and web technologies.

I love to solve problems in fields of data analysis, usability, user interaction and e-commerce. I enjoy working in small and dedicated teams of hackers.

Specialties:

* Search engines: finding faster and more accurate content. Search engines which learn from human interaction.
* Organic and paid search optimizations for e-commerce businesses.
* Usability improvements for e-commerce applications supported by data - A/B testing, MVT testing and advanced user behaviour tracking.
* Machine learning and data analysis.

Activity

Join now to see all activity

Experience

  • Adwords Robot Graphic

    Adwords Robot

    Enschede Area, Netherlands

  • -

    Enschede Area, Netherlands

  • -

    Enschede, Hengelo, Rotterdam

Education

  •  Graphic

    -

    -

    Activities and Societies: Arashi, Piranha Duiken, USE, VleugelLam, Euros Zeilen, InterActief, Scintilla, EHBO vereniging Enschede, CampusWIKI, SamenvattingenWIKI, Flatlijst, PicServer, CampusMagic, FileMe

    My master thesis was on automatic music transcription.

  • -

    -

    We worked on machine learning with applications in: smart conference rooms, multiple person video tracking and face recognition in security cameras.

  • -

    -

  • -

    -

    Activities and Societies: Arashi, Piranha, Vleugellam, Squash, Hiking & Survival, PicServer, CampusSearch, Flatlijst

Publications

  • Automatic Music Transcriber - an extendible framework

    Twente University

    We propose an open extensible framework for automatic music transcription (AMT) based on black board architecture. A basic AMT is implemented with a pitch detector and an onset detector modules. The framework is capable of containing low level sensory data as well high level knowledge about music, such
    as chord structures and meter information.

    Several transcription problems are addressed in detail. These include octave ambiguity errors, duration estimation errors and splitting of…

    We propose an open extensible framework for automatic music transcription (AMT) based on black board architecture. A basic AMT is implemented with a pitch detector and an onset detector modules. The framework is capable of containing low level sensory data as well high level knowledge about music, such
    as chord structures and meter information.

    Several transcription problems are addressed in detail. These include octave ambiguity errors, duration estimation errors and splitting of notes during multiple subsequent onsets. We propose a solution to some extent for all three problems. We also include analysis of harmonic properties of notes for live pianos and piano synthesizers.

    Our system is tested on a wide array of data, both synthesized MIDI ?les and live recorded sounds with polyphony level up to 8. We achieved an overall performance of 0.45 (F-measure), measured using three properties of notes: pitch, duration and onset time.

    See publication

Projects

  • FaboShop: Facebook Shop platform

    DUTCH: FaboShop is (waarschijnlijk) het eerste Facebook shopping platform in Nederland. Hierbij kunnen webwinkelier zonder maatwerk een eigen Facebook shop beginnen en Facebook als verkoopkanaal gaan gebruiken.

    Onze klanten: Prenatal.nl, DE.nl en jeanscentre.nl

    See project

Languages

  • Dutch

    Native or bilingual proficiency

  • English

    Full professional proficiency

  • Russian

    Native or bilingual proficiency

  • German

    Limited working proficiency

Recommendations received

6 people have recommended Dennis Join now to view

View Dennis’ full profile

  • See who you know in common
  • Get introduced
  • Contact Dennis directly
Join to view full profile

People also viewed

Explore collaborative articles

We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.

Explore More

Add new skills with these courses