Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM69040
Volume:
10 |
11 |
Nov 2026
Pages:
1-6
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
Department Of Computer Science and Engineering
K S Institute Of Technology
Department Of Computer Science and Engineering
K S Institute Of Technology
Department Of Computer Science and Engineering
K S Institute Of Technology
Department Of Computer Science and Engineering
K S Institute Of Technology
Department Of Computer Science and Engineering
K S Institute Of Technology
Commercial-space marketplaces depend on knowing how likely a listing is to be occupied and on presenting it to the right customers, yet a new platform starts without a long booking history. This paper presents SeekMySpace, a marketplace intelligence platform for commercial spaces in Bengaluru that addresses this cold-start problem with a source-tagged dataset and two AI modules. The forecasting module is a Random Forest regressor that predicts listing occupancy from nine features covering price, competitor price, footfall, booking count, location, category, events, holidays and customer segment. It is built on 199 records, 99 manually collected and 100 synthetic, of which 197 were usable, and achieves a mean absolute error of 0.0310 and an R² of 0.8194 on 40 held-out records. Each prediction reaches the owner with a demand score, a demand level and a suggested price. A supporting language-model module generates segment-specific, keyword-rich listing descriptions that remain tailored across property categories and customer segments. Both modules run behind a FastAPI service used by a React front end. By labelling every record as real, verified or synthetic, the approach gives early-stage platforms a transparent and reproducible way to start forecasting demand before large booking histories exist.
KEYWORDS
This paper described how SeekMySpace approached occupancy forecasting and listing generation for Bengaluru’s commercial spaces without a long booking history. The datasets keep every record labelled as real, verified or synthetic. A Random Forest predicts occupancy from nine features, with event information gathered for a seven-day window, and it reached an MAE of 0.0310,an RMSE of 0.0413 and an R² of 0.8194on 40 held-out records. A grounded prompt design lets a language model write audience-specific listing content without adding facts, as a worked example showed. The results are early and rest on partly synthetic data, but the approach offers a practical and transparent starting point for marketplaces that have listings before they have history.
[1] T. Somasekhar, S. H. Aditi, S. Bhavani, M. H. Bhoomika, and G. Brinda, “AI-driven marketplace intelligence for commercial spaces,” Int. J. Research and Scientific Innovation (IJRSI), vol. 13, no. 5, pp. 809–814, May 2026, doi: 10.51244/IJRSI.2026.1305000076. [2] N. T. Singh, M. Wadhwa, S. Kumar, V. M. Shrimal, U. Shrivastava, and Himanshu, “A study of the evolution of the real-estate flats and apartment rental system,” in Proc. 5th Int. Conf. Smart Electronics and Communication (ICOSEC), 2024, pp. 1846–1851, doi: 10.1109/ICOSEC61587.2024.10722232. [3] J. M. Shanthi, P. S. Rao, V. R. Kumar, G. Nithya, J. Kartigeyan, and A. Gupta, “Ensemble machine learning models used for forecasting in rental apartment price prediction of metro city in India,” in Proc. 6th Int. Conf. Emerging Technology (INCET),2025, doi: 10.1109/INCET64471.2025.11140248. [4] H. V. Singh, S. K. Srivastava, N. K. Pandey, and N. Kumar, “Integrating Random Forest and XGBoost for accurate house rent prediction,” in Proc. Int. Conf. Pervasive Computational Technologies (ICPCT), 2025, pp. 956–957. [5] L. Pallavi, D. Rishika, G. B. S. Sree, J. Ruthwik, and J. Yashaswini, “SmartRent: AI-driven online house rental platform,” in Proc. 5th Int. Conf. Soft Computing for Security Applications (ICSCSA), 2025, pp. 346–353, doi: 10.1109/ICSCSA66339.2025.11170770. [6] F. Shareef, “Enhancing conversational AI with LLMs for customer support automation,” in Proc. 2nd Int. Conf. Self Sustainable Artificial Intelligence Systems (ICSSAS), 2024, pp. 239–244, doi: 10.1109/ICSSAS64001.2024.10760403. [7] L. Breiman, “Random forests,” Machine Learning, vol. 45, no. 1, pp. 5–32, 2001. [8] F. Pedregosa et al., “Scikit-learn: Machine learning in Python,” J. Mach. Learn. Res., vol. 12, pp. 2825–2830, 2011.
Publication Date
11/18/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
2
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...