Construction and Inference Method of Semantic-Driven, Spatio-Temporal Derivation Relationship Network for Place Names
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThis manuscript proposed the methods for determining, representing, and using the spatiotemporal derivation relationships of place names. It is an interesting topic. Some suggestions are as follows.
References are crucial to support some statements, such as the following sentences, "Among them, Place names can be categorized into primary and related names based on their interrelationships", "The derivation relationship can be further divided into inheritance derivation, influence derivation, and spatiotemporal derivation", "The application of geographic name information can be traced back to the 17th century in China and the 19th century in the United Kingdom", "Scholars have conducted research on place name derivation and place name semantics from various perspectives, yet there is a lack of discussion on the spatiotemporal derivation relationships of place names in existing studies." etc. Providing references will strengthen the validity of these statements.
The authors defined the spatiotemporal derivation relationships as "when naming newly discovered geographical entities, people often combine the existing place names of surrounding natural or artificial geographical entities and generate new place names through the derivation of these existing place names." The definition and constraints of these relationships mainly focus on the spatial relationship. How about the temporal relationship?
Introduction Section
The Introduction Section could benefit from a more concise structure. Currently, it is not easy to follow the train of thought. Consider dividing this section into Introduction and Related Work to improve coherence and readability.
Section 2
This section introduced the concept, definition, and determination of the spatiotemporal derivation relationships of the place names. However, the title only reflects the first two parts.
Section 2.3 mainly introduces the methods for determining the derivative relationship of place names. The evaluation contents in this section can be moved to Section 5.2.
Section 5
Section 5.3 compared the spatiotemporal derivative relationship network of place names with Google Maps and Baidu Map. However, they are different. The network of place names only contains places with spatiotemporal derivation relationships. Baidu and Google store nearly all the place names and their locations. They can perform accurate queries when the target place name is located. So, please consider whether the conclusion "This approach enhances the accuracy and practicality of the retrieval results" is appropriate. Moreover, please consider whether Section 5.3 is necessary.
Comments on the Quality of English LanguageNo comments.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Comments and Suggestions for AuthorsVery good paper at the intersection of several disciplines: geography, linguistics, knowledge representation and reasoning, machine learning. The authors demonstrate the interest of formalizing and reasoning with the 'cognitive' relations that exist between place names, namely between 'primary place names' and 'secondary place names' that are linguistically and practically derived from them. The authors name these relationships: Spatiotemporal Derivation Relationships. In this paper they discuss the importance of considering such relationships for a number of qualitative reasoning functions related to geogaphic names and locations.
In the first section the authors present the motivations of the study and the interest of considering derivation relations between place names. They provide adequate mentions to previous works and showsome of their limits. They also emphasize the interest of considering the semantics of place names. They show that spatiotemporal derivation relationships of place names (I will abbreviate as STSRPN in this review) have not been studied and formalized and exploited in previous research.
In Section 2 the authors provide a rigorous definition of spatiotemporal derivation relationships of place names. They also clearly define and formalize the constraints (semantic and spatial) that hold on such relationships. Based on these constraints, the authors propose suitable methods that are used to identify STSRPN: semantic similarity, use of category ontologies, spatial constraint judgement based on topological relationships and spatial measurement.
In Section 3 the authors propose a method to build a spatiotemporal derivation network of place names that is used to record and reason about the derivation relationships between place names. They used the Neo4J graph database to build and exploit this network to develop a number of reasoning functions using spatial adjacency relationships. The paper explains how the Neo4J network is built and used to record the place names (primary and secondary) and the derivation relationships.
In Section 4 the authors present a number of inference functions that can be developed using the Neo4J network, taking advantage of the STSRPN: 1) inference of spatial relationships (going beyond the quantitative measures used in current software like Google Maps and Baidu Maps); 2) reasoning about spatial neighborhood; 3) inference of fuzzy spatial positions
In Section 5 experiments and results are presented to illustrate all the aspects discussed in the previous sections. The authors used Canadian place name data publicly obtained from OpenStreetMap. They show how they build the Neo4J network and how it is exploited. They provide comparisons with queries in Google Maps and Baidu Maps. It is shown how the proposed approach can be used to improve the results of these commercial services.
Hence, using this network, inference of spatial adjacency relationships provides an effective approach to enhance the expression of place name semantics, the retrieval of geographic information. It also enhances the accuracy of existing quantitative query results, and the reasoning of spatial locations provides a solution for data not yet included in the repository.
During the evaluation phase of the project it appeared that the approach failed in some cases: the use of abbreviations in place names, non standard ways to write place names, and limits of the decision tree developed by the authors. This leaves some opportunities to improve the proposed approach and associated software.
Very Good Job!
Comments on the Quality of English Language
The paper is very well written. Some corrections and precisions would be welcome
PAGE 2 PLEASE CHANGE THE SENTENCE: The above research carries out research on the derivation of place names
PAGE 5 CHANGE semantic con-straints are analyze BY semantic con-straints are analyzed
PAGE 5 IN the derived proper noun 𝑠𝑠𝑑 IT SEEMS THAT 𝑠𝑠𝑑 HAS NOT BEEN DEFINED
PAGE 6 YOU USE THE TERMS spatial constraint distance I WONDER IF IT SHOULD BE WRITTEN spatial distance constraint
PAGE 11 SECTION 3.2: THE DESCRIPTIONS OF THE PROPOSED 2 STEPS ARE VERY DIFFICULT TO FOLLOW, PLEASE PRESENT IT IN THE FORM OF AN ALGORITHM (INDENTED FORM)
PAGE 13 CHANGE the geographic entities and constraint information obtained from the element extraction process are reasoned according to the inference rules BY inference rules are used to reason about the geographic entities and constraint information obtained from the element extraction process
EVERYWHERE IT IS APPROPRIATE, PLEASE CHANGE What schools are near Fish Creek? BY Which schools are near Fish Creek?
PAGE 14 CHANGE with the geographical entity to be reasoned BY with the considered geographical entity
PAGE 14 IN The process of the method in this paper is shown in the figure, PLEASE INDICATE WHICH FIGURE
PAGE 21 IT SEEMS THAT FIGURE 11 DOES NOT CORRESPOND TO THE TEXT
REFERENCE LIST:
REFERENCES 26 AND 27 ARE NOT PRECISE ENOUGH
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 3 Report
Comments and Suggestions for AuthorsThis paper studies the spatiotemporal derivation relationship of place names and defines the spatiotemporal semantic network of place names. The method of this paper has certain application value for the diversification of place names retrieval services.
The main recommendations of the paper are as follows:
1. Equations (1) and (2) do not reflect the spatio-temporal relationship.
2. Equation (3) is a bit far-fetched to define semantic similarity.
3. The method adopted in the paper is relatively simple, and the innovation is not enough.
4. The first application case in Section 5.3.1 does not reflect the advantages of the proposed method, which should be used to address the limitations of defining geographical name retrieval in existing map services.
5. The experiments in this paper are only manually verified by limited data, and are not compared and analyzed with other related methods, which is not convincing enough.
6. There are some spelling mistakes, such as "spatiotemporal" suggesting a unified "spatio-temporal".
Comments on the Quality of English LanguageThere are some spelling mistakes, such as "spatiotemporal" suggesting a unified "spatio-temporal".
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 3 Report
Comments and Suggestions for AuthorsSorry, judging by the revised manuscript, I don't think it's enough to accept the publication of the paper's work. I also think that the paper is not innovative enough, and there are still some typographical errors. It is recommended to apply to other journals that are more suitable.
Comments on the Quality of English LanguageThere are still some typographical errors. For example, p11, Some Chinese words are in the algorithm description. It is recommended to apply to other journals that are more suitable.
