Search
Generic filters
Exact matches only
Filter by content type
Users
Attachments

Lehre

Überblick

Das Hauptziel unserer Lehrveranstaltungen ist es, die Datenkompetenz der Studierenden zu erhöhen. Wir verfolgen einen problem-orientierten Fallstudienansatz, der es Studierenden ermöglicht, herausfordernde betriebswirtschaftliche Probleme in datenbezogene Fragestellungen zu übersetzen und diese mit Ansätzen aus den Bereichen Business Analytics, Data Science und Machine Learning zu lösen sowie die erzielten Ergebnisse kritisch zu hinterfragen.

Kurse HS22 & FS22

Bachelor
Master
PhD
add

Business Analytics und Data Science Applications

Ivo Blohm

Daten sind das neue Öl! Daten sind eine neue Asset-Klasse! Solche und ähnliche Aussagen sind heute allgegenwärtig und sollen das grosse wirtschaftliche Potenzial von Data Science und künstlicher Intelligenz veranschaulichen. "Smarte" Produkte und Dienstleistungen sind bereits heute allgegenwärtig und werden in Zukunft eine immer grössere Rolle in unserem Privat- und Arbeitsleben spielen. So galten beispielsweise selbstfahrende Autos vor 10 Jahren noch als technische Utopie; heute wartet man auf die flächendeckende Markteinführung. Übergeordnetes Ziel des Kurses ist es, Studierende in die Lage zu versetzen, mittels Big Data, Data Science, Machine Learning & Co betriebswirtschaftliche Probleme lösen zu können. Dafür verfolgt der Kurs ein Fallstudien-orientiertes Kursdesign. In jeder Fallstudie wird auf Basis einer konkreten betriebswirtschaftlichen Problemstellung eine prototypische Anwendung entwickelt, um z.B. die finanzielle Performance, operative Effizienz oder die Effektivität von betriebswirtschaftlichen Massnahmen zu verbessern. Im Rahmen der Fallstudien werden eine Vielzahl von unterschiedlichen Algorithmen und Analyse-Ansätzen besprochen und eingeübt. Die Fallstudien stammen dabei in der Regel aus dem Kontext von führenden Internet- und Technologie-Unternehmen, wie z.B. Facebook, Linkedin, Netflix, Orange oder Zalando. Diese Fallbeispiele umfassen dabei u.a. die Vorhersage von Weinqualität, Vorhersage von zukünftigen Kundenverhalten, das Erstellen von Kaufempfehlungen oder die automatisierte Identifikation von Fake News. Studierende lernen dabei den Umgang mit grossen unstrukturierten Datenmengen.

Abschlussarbeiten

Sind Sie am Ende Ihres Bachelor- oder Masterstudiums? Im Rahmen unserer Forschungsschwerpunkte suchen wir fortlaufend Studierende, die bei uns eine Abschlussarbeit schreiben wollen.

Level:
Bachelor/Master
Kontakt:
Philipp Gordetzki

Why:
Digital innovation platforms are popular solutions for engaging consumers in the innovation process for several reasons: faster innovation, higher quality ideas, and lower costs are just some of the benefits. Nevertheless, there is currently still a need for improvement. Ideas are developed and evaluated by people like you and me, who tend to suffer from barriers to creativity, such as cognitive fixation, aversion to novelty, and reluctance to leave the path of least resistance.
After the ideation process, evaluating a large number of ideas is a difficult task for human experts. AI can solve this problem by automatically analyzing ideas with regard to their novelty, degree of elaboration, and innovativeness. Your task is to investigate how an AI can analyse existing feedback and provide guidance during evaluation. We provide you with an extensive real-world dataset containing over 100,000 ideas with comments.

How?

• Analyse the comments in the dataset to understand how humans provide feedback
• Utilize modern technologies such as BERT or GPT to train an AI to learn how to provide feedback
• Optionally apply your solution in a real-world experiment to test your solution

Key-Facts:

• Close supervision, with regular review meetings, feedback discussions, etc.
• Thesis can start immediately but should be completed within the next ±6 months
• If the work is worth publishing, you will be listed as an author

If you are interested, send your CV, transcript of records, and a brief description of your motivation to Philipp Gordetzki (). Looking forward to hearing from you! 🙂

Level:
Bachelor
Kontakt:
Philipp Gordetzki

Why:
Digital innovation platforms are popular solutions for engaging consumers in the innovation process for several reasons: faster innovation, higher quality ideas, and lower costs are just some of the benefits. Nevertheless, there is currently still a need for improvement. Ideas are developed and evaluated by people like you and me, who tend to suffer from barriers to creativity, such as cognitive fixation, aversion to novelty, and reluctance to leave the path of least resistance.
Your task is to identify how AI can be utilized on innovation platforms. You identify previous use-cases and describe challenges to guide further research.

How?

• Thesis that conducts a systematic literature review on use cases of AI on innovation platforms using the methodology of Webster and Watson (2002) and Vom Brocke et al. (2009; 2015)

Key-Facts:

• Close supervision, with regular review meetings, feedback discussions, etc.
• Thesis can start immediately but should be completed within the next ±6 months
• If the work is worth publishing, you will be listed as an author

If you are interested, send your CV, transcript of records, and a brief description of your motivation to Philipp Gordetzki (). Looking forward to hearing from you! 🙂

Level:
Bachelor/Master
Kontakt:
Philipp Gordetzki

Why:
Digital innovation platforms are popular solutions for engaging consumers in the innovation process for several reasons: faster innovation, higher quality ideas, and lower costs are just some of the benefits. Nevertheless, there is currently still a need for improvement. Ideas are developed and evaluated by people like you and me, who tend to suffer from barriers to creativity, such as cognitive fixation, aversion to novelty, and reluctance to leave the path of least resistance.
During the ideation process, feedback is crucial but costly for human editors. AI can solve this problem by automatically providing high-quality feedback to enhance human creativity. Your task is to investigate how an AI can learn from human feedback. We provide you with an extensive real-world dataset containing over 100,000 ideas with comments.

How?

• Analyse the comments in the dataset to understand how humans provide feedback
• Utilize modern technologies such as BERT or GPT to train an AI to learn how to provide feedback
• Optionally apply your solution in a real-world experiment to test your solution

Key-Facts:

• Close supervision, with regular review meetings, feedback discussions, etc.
• Thesis can start immediately but should be completed within the next ±6 months
• If the work is worth publishing, you will be listed as an author

If you are interested, send your CV, transcript of records, and a brief description of your motivation to Philipp Gordetzki (). Looking forward to hearing from you! 🙂

Level:
Bachelor/Master
Kontakt:
Philipp Gordetzki

Why:
Digital innovation platforms are popular solutions for engaging consumers in the innovation process for several reasons: faster innovation, higher quality ideas, and lower costs are just some of the benefits. Nevertheless, there is currently still a need for improvement. Ideas are developed and evaluated by people like you and me, who tend to suffer from barriers to creativity, such as cognitive fixation, aversion to novelty, and reluctance to leave the path of least resistance.
Understanding what makes an idea new and exciting is crucial for generating and evaluating innovation. AI can solve this problem by automatically predicting the novelty of an idea. Your task is to investigate existing innovation ideas and apply text mining to assess originality. Next, an AI can learn the patterns behind novelty. We provide you with an extensive real-world dataset containing over 100,000 ideas with comments.

How?

• Analyse the ideas in the dataset and find patterns and clusters
• Utilize text mining to generate measurements for novelty and conceptualize a novelty score that an AI can predict
• Optionally apply your solution in a real-world experiment to test your solution

Key-Facts:

• Close supervision, with regular review meetings, feedback discussions, etc.
• Thesis can start immediately but should be completed within the next ±6 months
• If the work is worth publishing, you will be listed as an author

If you are interested, send your CV, transcript of records, and a brief description of your motivation to Philipp Gordetzki (). Looking forward to hearing from you! 🙂

Level:
Bachelor/Master
Kontakt:
Philipp Gordetzki

Why:
Digital innovation platforms are popular solutions for engaging consumers in the innovation process for several reasons: faster innovation, higher quality ideas, and lower costs are just some of the benefits. Nevertheless, there is currently still a need for improvement. Ideas are developed and evaluated by people like you and me, who tend to suffer from barriers to creativity, such as cognitive fixation, aversion to novelty, and reluctance to leave the path of least resistance.
AI can help users conceptualize an innovative idea by providing automatic feedback or cognitive stimulation. Different AI-supported writing methods exist today, e.g., spelling checks such as Grammarly. Your task is to create an AI-supported user journey on an innovation platform. You can build an experiment with our team to test your hypotheses and compare different methods.

How?

• Conceptualize a user journey with different support opportunities for an AI
• Implement prototypes of such support opportunities in our experiment platform
• Test your solution in a real-world experiment

Key-Facts:

• Close supervision, with regular review meetings, feedback discussions, etc.
• Thesis can start immediately but should be completed within the next ±6 months
• If the work is worth publishing, you will be listed as an author

If you are interested, send your CV, transcript of records, and a brief description of your motivation to Philipp Gordetzki (). Looking forward to hearing from you! 🙂

Level:
Bachelor
Kontakt:
Philipp Gordetzki

Why:
Digital innovation platforms are popular solutions for engaging consumers in the innovation process for several reasons: faster innovation, higher quality ideas, and lower costs are just some of the benefits. Nevertheless, there is currently still a need for improvement. Ideas are developed and evaluated by people like you and me, who tend to suffer from barriers to creativity, such as cognitive fixation, aversion to novelty, and reluctance to leave the path of least resistance.
AI can help users to overcome these barriers. In the past, AI was successfully used in various domains of human creativity, e.g., creating music, images, or novels. As innovation is a challenging creativity process, we aim to support humans in a hybrid approach. Your task is to investigate previous use-cases of AI to increase human creativity and identify gaps and opportunities for further research.

How?

• Thesis that conducts a systematic literature review on the topic of hybrid creativity using the methodology of Webster and Watson (2002) and Vom Brocke et al. (2009; 2015)

Key-Facts:

• Close supervision, with regular review meetings, feedback discussions, etc.
• Thesis can start immediately but should be completed within the next ±6 months
• If the work is worth publishing, you will be listed as an author

If you are interested, send your CV, transcript of records, and a brief description of your motivation to Philipp Gordetzki (). Looking forward to hearing from you! 🙂

Haben Sie einen eigenen Themenvorschlag?

Haben Sie ein eigenes Thema, das für uns interessant sein könnte? Wir freuen uns über Ihren Themenvorschlag!

    Ihre Bewerbung für eine Abschlussarbeit

    Danke für Ihr Interesse, Ihre Abschlussarbeit bei Prof. Dr. Blohm zu schreiben. Wir betreuen gerne Studierende, die sich für unsere Forschungsthemen begeistern und die bisher gute akademische Leistungen erbracht haben. Weil wir eine grosse Anzahl an Bewerbungen bekommen, können wir nur jene berücksichtigen, die gut zu unseren Forschungsgebieten passen. Mehr zu unseren Forschungsgebieten erfahren Sie auf dieser Webseite. Um sich zu bewerben, können Sie eines der Themen in Betracht ziehen, an denen wir arbeiten oder Sie können ein eigenes Thema vorschlagen.



    Bewerbung für:*

    Research Proposal:*

    CV:*

    Motivationsschreiben:*

    Notenauszug:

    Vorgesehener Beginn der Arbeit:*

    Vorgesehene Abgabe der Arbeit:*
    Wir rechnen mit mindestens 4 Monaten Bearbeitungszeit von Beginn bis zur Abgabe der Arbeit