Using Natural Language Processing Techniques to Tackle the Construct Identity Problem in Information Systems Research
Research output: Contributions to collected editions/works › Article in conference proceedings › Research › peer-review
Standard
Proceedings of the 53rd Annual Hawaii International Conference on System Sciences, HICSS 2020: Knowing What We Know: Theory, Meta-analysis, and Review. ed. / Tung X. Bui. Honolulu: University of Hawaiʻi at Mānoa, 2020. p. 5675-5684 (Proceedings of the Annual Hawaii International Conference on System Sciences; Vol. 2020-January).
Research output: Contributions to collected editions/works › Article in conference proceedings › Research › peer-review
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - CHAP
T1 - Using Natural Language Processing Techniques to Tackle the Construct Identity Problem in Information Systems Research
AU - Ludwig, Siegfried
AU - Funk, Burkhardt
AU - Mueller, Benjamin
N1 - Conference code: 53
PY - 2020/1/7
Y1 - 2020/1/7
N2 - The growing number of constructs in behavioral research presents a problem to theory integration, since constructs cannot clearly be discriminated from each other. Recently there have been efforts to employ natural language processing techniques to tackle the construct identity problem. This paper compares the performance of the novel word-embedding model GloVe and different document projection methods with a latent semantic analysis (LSA) used in recent literature. The results show that making use of an advantage in document projection that LSA has over GloVe, performance can be improved. Even against this advantage of LSA, GloVe reaches comparable performance, and adjusted word embedding models can make up for this advantage. The proposed approach therefore presents a promising pathway for theory integration in behavioral research.
AB - The growing number of constructs in behavioral research presents a problem to theory integration, since constructs cannot clearly be discriminated from each other. Recently there have been efforts to employ natural language processing techniques to tackle the construct identity problem. This paper compares the performance of the novel word-embedding model GloVe and different document projection methods with a latent semantic analysis (LSA) used in recent literature. The results show that making use of an advantage in document projection that LSA has over GloVe, performance can be improved. Even against this advantage of LSA, GloVe reaches comparable performance, and adjusted word embedding models can make up for this advantage. The proposed approach therefore presents a promising pathway for theory integration in behavioral research.
KW - Informatics
KW - construct identity fallacy
KW - global vectors for word representation (glove)
KW - jingle jangle
KW - latent semantic analysis (lsa)
KW - word embeddings
UR - http://www.scopus.com/inward/record.url?scp=85082300548&partnerID=8YFLogxK
UR - https://hdl.handle.net/10125/64069
UR - https://www.mendeley.com/catalogue/cf195722-124c-37d7-853c-fc8a935dcfdb/
M3 - Article in conference proceedings
T3 - Proceedings of the Annual Hawaii International Conference on System Sciences
SP - 5675
EP - 5684
BT - Proceedings of the 53rd Annual Hawaii International Conference on System Sciences, HICSS 2020
A2 - Bui, Tung X.
PB - University of Hawaiʻi at Mānoa
CY - Honolulu
T2 - Hawaii International Conference on System Sciences - HICSS 2020
Y2 - 7 January 2020 through 10 January 2020
ER -