Collaborative investing platforms often rely on the common “wisdom of the crowd” concept in a domain in which even highly paid, well-educated, and experienced professionals make mistakes, providing wrong or inaccurate evaluations.

Under the realistic assumption of the coexistence between capable experts and average workers, the authors analyze data from two well-known platforms, focusing on the thin line that separates the direct and the indirect wisdom of crowds. Indeed, the empirical experiments, consisting in the correlation between sentiment analysis and historical performance of relevant stocks, clearly show that experts can be identified. Those people provide the real key value. The remaining part of the network plays a key role by helping to indirectly identify the valuable content through interactions.

I appreciated this contribution. The authors provide clear results and seem aware of both the advantages as well as the limitations of the adopted approach. Moreover, the future work outlined in the paper looks very promising, especially concerning the possible implementation of systems underpinned by meta-reputation.