Next Article in Journal
Event Geoparser with Pseudo-Location Entity Identification and Numerical Argument Extraction Implementation and Evaluation in Indonesian News Domain
Next Article in Special Issue
The Architecture of Mass Customization-Social Internet of Things System: Current Research Profile
Previous Article in Journal
Supporting SDG 15, Life on Land: Identifying the Main Drivers of Land Degradation in Honghe Prefecture, China, between 2005 and 2015
Article
Peer-Review Record

OurPlaces: Cross-Cultural Crowdsourcing Platform for Location Recommendation Services

ISPRS Int. J. Geo-Inf. 2020, 9(12), 711; https://doi.org/10.3390/ijgi9120711
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Reviewer 3: Anonymous
Reviewer 4: Anonymous
ISPRS Int. J. Geo-Inf. 2020, 9(12), 711; https://doi.org/10.3390/ijgi9120711
Received: 14 October 2020 / Revised: 12 November 2020 / Accepted: 13 November 2020 / Published: 27 November 2020
(This article belongs to the Special Issue Intelligent Systems Based on Open and Crowdsourced Location Data)

Round 1

Reviewer 1 Report

Authors propose a cross-cultural crowdsourcing platform to share the people opinion of different cultures visiting cities. The paper is interesting, it is well written, but it is necessary a review in some sections to do more easy to read. The formulation and experimental sections are very complete.

Suggestions:

- The introduction section is confused and hard to read. I recommend to highlight your main contributions.

-As a personal recommendation, the dataset has to be pre-filtered to remove the biased comments, or have to include a margin of false bad-comments

- Could be interesting to include other dataset from similar apps.

Author Response

Thank you for your careful reviews. We have made every attempt to fully address these comments in the revised manuscript. In enclose file, we outline how we have handled comments of each reviewer. Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The topic is interesting for the research community. Moreover, the paper is well-written and organized.

However, some references are too out of dated. It is suggested to update the references to the recent years.

The paper lacks a discussion and elicitation of findings from the existing works.

Conclusions seem like an abstract - it is suggested to present more general results here and moreover to mention potential future research.

Author Response

Thank you for your careful reviews. We have made every attempt to fully address these comments in the revised manuscript. In enclose file, we outline how we have handled comments of each reviewer. Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

This paper proposes a cross-cultural crowdsourcing platform where people from different cultures can share their spatial experiences. And then the author presents a cross-cultural tourism recommendation system based on the cognitive similarity between users. Experiments on a real dataset demonstrate the effects of the cross-cultural tourism recommendation system. The proposed idea is originality and solves the practical problems. However, it should still go through careful revision by considering the following comments.

 

  1. The definition of cognition is not clear, the “cognition” in this paper looks like the aggregation of some locations. However, user cognition describes a psychological process, including feeling, perception, memory, thinking, imagination, and language, etc. It focuses on the process of the human brain's processing of external information, and influences user decisions based on external information. Although the authors have repeatedly emphasized that the core and innovative point is cognition, the current text does not reflect the original meaning of cognition.

 

  1. There is an insufficient summary of the latest and advanced location recommendation methods. The related work needs to be expanded and include some of the state-of-the-art in the area (most of the references are old). Collaborative filtering method has derived many advanced variants, such as the neural collaborative filtering method combined with deep learning. The authors should introduce these research progresses.

 

  1. The description of the proposed method is inadequate. For example, the classification of user groups is unclear in section 3.1, and the computation of is unclear in Equation (4). I suggest the authors give them in details. A notation table should be given to facilitate the understanding of the paper. 

 

  1. The experiments are only conducted on a small-scale dataset with 50 users and 2000 cognitive feedbacks. The large-scale data should be evaluated to demonstrate the performance. 

 

  1. Using only MAE cannot comprehensively evaluate recommendation performance. RMSE and MAE evaluate the recommendation performance from different perspectives. Generally speaking, these two criteria should be considered simultaneously. I recommend the authors to update their experiments.

 

  1. The subsection 5.3 lacks the necessary discussion and analysis of experimental results. I suggest the author expand them carefully. The parameter that is the priority of users in selecting similar tourism places plays an important role in the proposed method. I suggest the authors conduct experiments to study the effect of the  

 

  1. I suggest the authors expand the discussion and conclusion carefully.

 

  1. The paper is comprised of many grammar mistakes and unprofessional presentations. I suggest the authors carefully rewrite the paper before re-submission. 

Author Response

Thank you for your careful reviews. We have made every attempt to fully address these comments in the revised manuscript. In enclose file, we outline how we have handled comments of each reviewer. Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

The topic of the paper is very interesting, resulting in some new tools for tourism industry. The literature review and methodology are well written, but there are some limitations that need to be taken into consideration:

  • The first-person pronouns as “we” or “our” should be avoided in order to maintain the tertiary nature of the publication and a neutral voice in the article;
  • Based on what did the authors conclude that CF has problems like cold-start and sparsity (line 34-35 and the same in the line 86)?
  • Text in lines 46-72 is more appropriate for Methods, but since this section is in details explained, this part can be excluded for the introduction.
  • Text in lines 73-78 is not necessary.
  • I suggest that the title of the third section (Crowdsourcing Platform for Measuring Cognitive Similarity) be changed to “Materials and Methods”.
  • Text in lines 226-233 is just repeated from the previous section, so it should be left out or changed.
  • Although authors explained in details the procedure of building a three layered architecture, the discussion is missing and the conclusion section is quite week (text in this section is mostly repeated from the previous sections). I recommend adding practical implications, recommendations for future research and limitations of the proposed model.

Author Response

Thank you for your careful reviews. We have made every attempt to fully address these comments in the revised manuscript. In enclose file, we outline how we have handled comments of each reviewer. Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

The authors have rewritten the paper and most of the revisions followed the suggestions. Therefore, I recommend to accept the paper after minor revisions. 

Author Response

Thank you for your comment, please see the attached file

Author Response File: Author Response.pdf

Back to TopTop