Refine
Document Type
- Conference Proceeding (2)
- Report (1)
Language
- English (3)
Keywords
- Artificial Intelligence (3) (remove)
Text produced by entrepreneurs represents a data source in entrepreneurship research on venture performance and fund-raising success. Manual text coding of single variables is increasingly assisted or replaced by computer-aided text analysis. Yet, for the development of prediction models with several variables, such dictionary-based text analysis methods are less suitable. Natural language processing techniques are an alternative; however, the implementation is more complex and requires substantial programming skills. More work is required to understand how text analytics can advance entrepreneurship research. This study hence experiments with different artificial intelligence methods rooted in Natural Language Processing and deep learning. It uses 766 business plans to train a model for the automated measurement of transaction relations, a construct which is an indicator for new technology-based firm survival. Empirical findings show that the accuracy of construct measurement can be significantly increased with automated methods and improves with larger amounts of training data. Language complexity sets limits to the precision of automated construct measurement though. We therefore recommend a hybrid approach: making use of the inherent advantages of combining automated with human coding until the amount of training data is sufficiently large to substitute the human coding completely. The study provides insights into the applicability of different text analytics methods in entrepreneurship research and points at future research potential.
The development of a new product can be accelerated by using an approach called crowdsourcing. The engineers compete and try their best to provide the related solution based on the given product requirement submitted in the online crowdsourcing platform. The one who has submitted the best solution get a financial reward. This approach is proven to be three time faster than the conventional one. However, the crowdsourcing process is usually not transparent to a new user. The risk for the execution of a new project for developing a new product is not easy to be calculated [1, 2]. We developed a method InnoCrowd to handle this problem and the new user could use during the planning of a new product development project. This system uses AI concepts to generate a knowledgebase representing histories of successful product development projects. The system uses the knowledge to determine qualitative and quantitative risks of a new project. This paper describes the new method, the InnoCrowd design, and results of a validation experiment based on data from a current crowdsourcing platform. Finally, we compare InnoCrowd to related methods and systems in terms of design and benefits.
Research Report
(2024)