New to Magnet.me? Sign up now to get your own personalised job matches.

Sign up

Graduation project: Smart Search Engines for Chatbots

  • Type Internship
  • Degree Master
  • Location Zoetermeer, NetherlandsZoetermeerNL
  • Posted
  • Start date -
  • Deadline -
  • Full-time graduate internship
Find your job faster with a profile
Directly see your job matches and find a job that's perfect for you.

As a Graduate Intern at ORTEC you will be part of a unique organization and team! With this graduate project, you will be part of ORTEC’s student team in the Text Mining LAB. Together with other graduate students, you will work within the same strategic topic, but you will have your own research question and thesis subject. Within this LAB you will have the possibility to brainstorm and discuss with colleagues about the topic and everything that is encompassed in writing a thesis, leading to a thesis with more impact.

The aim of this Master Thesis is to create a smart search engine for a Chatbot. Chatbots consist of algorithms that can interpret and answer a variety of questions. Those algorithms are typically a combination of natural language processing (NLP), machine learning (mostly classification models) and smart business rules. A smart search engine can be a worthy addition to that list and has the potential to increase the answer ‘flexibility’ without requiring extra programming efforts. Smart search engines also have a downside. They can return unpredictable results and deliver ‘wrong’ answers. In this project we would like to investigate the use of some modern technologies to develop a new approach on creating smart search engines. For example:

A chatbot uses a search engine to go through a large set of FAQs. A user asks a question about topic X and the search engine finds 8 relevant articles about that topic. A smart search engine is capable of selecting the right (= most relevant) article based on the original question. 

In this search process, two steps are important:

  1. Finding relevant articles based on keywords. This step uses a combination of NLP, machine learning and information retrieval algorithms (i.e. doc2vec);
  2. Classification of output. Named entity recognition and topic modelling could be important tools in this step, in which we determine which article we return to the user from the set of articles that were identified as relevant.

In this project we are also interested in the comparison of some ‘off the shelf’ algorithms such as IBM Watson Discovery (possibly enhanced with Watson Knowledge Studio) and Microsoft Azure Search (possibly enhanced with Azure Text Analytics) and a self-made version based on ElasticSearch (combined with doc2vec).

Those different options can be compared based on the following criteria

  1. Performance of the algorithm
  2. Business case:

a. Required time investment of data scientist and business expert;

b. Required maintenance;

c. License fees.

If you are interested, we would like to receive your resume, motivation and grade list. You can send your application to recruitment@ortec.com.  As a student you can complete your master thesis and/or work part-time as a student assistant. If you want to know more about the possibilities or need more information, then please contact Gordon Boon (Recruiter).

Who you are

  • You are a student in the masters phase of your course in the area of Econometrics, Mathematics, Computer Science, Business Informatics, Information Science or comparable
  • You can handle a high degree of complexity and work in a team
  • You are passionate about IT – it’s no coincidence that this is your chosen field of study
  • You possess skills in coding algorithms, including for machine learning
  • Able to fluently communicate in English (verbally and in writing).

What we offer

  • A practical and challenging case in which you can exercise your theoretical knowledge
  • Inspiring, smart and enthusiastic colleagues
  • Ample opportunities to develop yourself (internationally)
  • Excellent supervision during your graduation project
  • Good internship allowance and a laptop
  • Activities with all graduate students at ORTEC, like Friday afternoon drinks
  • Possibility to work as a Student Assistant next to your graduation project

What to expect

We help you to thrive in your field of expertise. We operate a flat organizational structure that keeps communication lines short. The atmosphere is open, informal, cooperative and positive. We employ over 900 people in the Netherlands (HQ), Belgium, Germany, France, the U.K., Romania, Italy, the U.S., Australia, Brazil, Poland and Denmark.

Visit our website ortec.com to learn more about our solutions and clients' experiences.

Acquisition as a result of this vacancy is not appreciated.

People also viewed