Review Reports - Review on the Application of Artificial Intelligence in Smart Homes

Round 1

Reviewer 1 Report

I read this paper with a lot of interest. I find that a review paper on AI for smart homes is always needed and I believe that the paper has merit. What I liked in this paper is the simplicity with which the authors describe the artificial intelligence (its concept in smart homes) and the various proposed technologies. The paper is mainly for people outside of the field of engineering or compute science, and I believe it will be useful for those people. As such I would suggest that the paper get considered only after some changes that need to be done. In particular I suggest the following:

- I would suggest that the authors add a contribution in the introduction that clearly states that this paper is mainly focusing on people that are outside of engineering and computer science fields.

- The main problem of the paper is that there is no a discussion section. The discussion section should summarize the main conclusions of the paper and provide a criticism of the current state of the art of AI in smart homes. The authors based on their review have to say whether there is room for further improvements of AI in smart homes or not.

- A reference is needed after the sentence “The ideal state of artificial intelligence is “Thinking Humanly, Thinking Rationally, Acting Humanly, and Acting Rationally”. This is a phrase taken from Russell book on AI. Please cite it there (or any other source that contains that phrase).

- What do the rows with dots “…” in Table 1 mean? This is a sample table that provides some samples from the literature review, so I do not fully understand what sample is the “…”?

- Also the connection of the tables in the text with the table in the appendix should be further clarified. Why is this distinction?

- There are several syntactical and grammatical errors in the text. Overall it is at a good level but definitely a slight language polishing is needed. Some mistakes are given below:

o “It will analysis sensor data” should be “It will analyse sensor data”

o “home also need to be discussed” should be “home also needs to be discussed”

o The sentence “we next distributed AI technology in smart home into 6 clusters” reads awkward

o “is not so many” should be “is not so high”

o “AI for decisions making is more commonly” should be “AI for decision-making is more commonly”

o “predictions making” should be “prediction making”

Author Response

Dear Reviewer,

We are truly grateful to your critical comments and thoughtful suggestions. Based on these comments and suggestions, we have made careful modifications on the original manuscript. We hope the new manuscript will meet your standard. Below you will find our point-by-point responses to the comments:

Point 1: I would suggest that the authors add a contribution in the introduction that clearly states that this paper is mainly focusing on people that are outside of engineering and computer science fields.  

Response 1: We revised and explained our contribution in the introduction on page 3: “This research contributed to find out the trends of smart home technology and product and state the relationship between literature and product in this field. We conducted this review from the perspective of architecture and human concern.”

Point 2: The main problem of the paper is that there is no a discussion section. The discussion section should summarize the main conclusions of the paper and provide a criticism of the current state of the art of AI in smart homes. The authors based on their review have to say whether there is room for further improvements of AI in smart homes or not.

Response 2: We put Result section and Discussion section together but maybe we didn’t state it clearly. Therefore, we summarize our opinion and also point out our limitation on page 11: “Generally, there is room for further improvements of AI in smart home. Currently, smart home is more utilized in energy management, intelligent interaction and security with AI technologies of voice recognition and image recognition. In the foreseeable future, more and more products will use activity recognition, data processing and predictions making. There may be some possible limitations in this study. First, our subcategories for the AI was not be chosen in a very systematic way. Second, the smart home product databased we chosen doesn’t cover the newest products. Third, if the keywords used in this article do not appear in some relevant publications, they would not be searched.”

Point 3: A reference is needed after the sentence “The ideal state of artificial intelligence is “Thinking Humanly, Thinking Rationally, Acting Humanly, and Acting Rationally”. This is a phrase taken from Russell book on AI. Please cite it there (or any other source that contains that phrase). 

Response 3: We already cited the “Artificial Intelligence: A Modern Approach, 3rd ed” in our article.

Point 4: What do the rows with dots “…” in Table 1 mean? This is a sample table that provides some samples from the literature review, so I do not fully understand what sample is the “…”? Also the connection of the tables in the text with the table in the appendix should be further clarified. Why is this distinction?

Response 4: Because the table is too long, we use ellipses to indicate that some of the data is omitted from the "sample table." The tables in the text is part of the tables in the appendix.

Point 5: There are several syntactical and grammatical errors in the text. Overall it is at a good level but definitely a slight language polishing is needed. Some mistakes are given below:

l “It will analysis sensor data” should be “It will analyse sensor data”

l “home also need to be discussed” should be “home also needs to be discussed”

l The sentence “we next distributed AI technology in smart home into 6 clusters” reads awkward

l “is not so many” should be “is not so high”

l “AI for decisions making is more commonly” should be “AI for decision-making is more commonly”

l “predictions making” should be “prediction making”

Response 5: We already proofread our article and made some revision.

Author Response File: Author Response.docx

Reviewer 2 Report

The paper overviews the interrelations between artificial intelligence and smart homes, particularly analyzing the literature published in this areas, as well as the products on the market.

Such an overview is interesting, as in principle, both areas are on the rise and are actively evolving. Some interesting dependencies are also detected, as for example, that the interaction aspect is gaining importance.

The paper has though some shortcomings:

- The chosen subcategories for the artificial intelligence areas (e.g. data mining…) appear to be chosen in a very ad-hoc manner. They do not appear to cover the field as a whole. For example, many smart home applications and publications were made basing on the rule-based technologies and semantic systems. These techniques are also an important part of AI and should be considered. The works and publications in these areas are vast and numerous, and are often addressing the areas for which the literature survey here found too few examples (for example, in electrical appliance management). Some examples of such semantic smart home works include the works of Anna Fensel and colleagues, of Dario Bonino and colleagues, as well as relevant IoT ontology standardization efforts, such as SAREF.

Generally, a more systematic approach to an overview should be picked here, and executed throughout the paper.

- The paper, even though it is readable, has numerous English language mistakes, often very basic ones. E.g. in the 1^st sentence of the abstract “improve the ability” (of what?), the last line of the 1^st page: “systems plays” – a grammar mistake, on page 2: “for the sake of better understand” – not a fluent/correct expression, and so on.

Author Response

Dear Reviewer,

Point 1: The chosen subcategories for the artificial intelligence areas (e.g. data mining…) appear to be chosen in a very ad-hoc manner. They do not appear to cover the field as a whole. For example, many smart home applications and publications were made basing on the rule-based technologies and semantic systems. These techniques are also an important part of AI and should be considered. The works and publications in these areas are vast and numerous, and are often addressing the areas for which the literature survey here found too few examples (for example, in electrical appliance management). Some examples of such semantic smart home works include the works of Anna Fensel and colleagues, of Dario Bonino and colleagues, as well as relevant IoT ontology standardization efforts, such as SAREF. Generally, a more systematic approach to an overview should be picked here, and executed throughout the paper.

Response 1: We reviewed the works of Anna Fensel, Dario Bonino and colleagues, as well as Laura Daniele’s article about SAREF and then reconsidered the subcategories for the artificial intelligence areas. We changed “data mining” into “data processing”, which in this article include the concept of data mining, semantic analysis and rule-based technologies. But we think we should have a more systematic approach to chose the subcategories so we put this point in Limitation. You can find it on page 11.

Point 2: The paper, even though it is readable, has numerous English language mistakes, often very basic ones. E.g. in the 1st sentence of the abstract “improve the ability” (of what?), the last line of the 1st page: “systems plays” – a grammar mistake, on page 2: “for the sake of better understand” – not a fluent/correct expression, and so on. 

Response 2: We proofread our article and made some revision. E.g. in the 1^st sentence of the abstract: “improved the ability of providing occupants with better comfort”; The last line of the 1^st page: “In these systems, AI plays the role of a knowledge and rule database, decision maker, action implementor and appliance controller”;On page 2: “For easy understanding”.

Author Response File: Author Response.docx

Reviewer 3 Report

The authors present a survey on the Application of Artificial Intelligence in Smart Home. Reviewer has some major and critical suggestions for the paper,

Their and too many grammatical errors starting from the abstract. Missing punctuations and other Typos is degrading the readability of the manuscript. Please revise the manuscript with appropriate proof reading.

Reviewer is unable to understand the novelty of the work. The outcome of the survey is specified as "our results suggest the relationship between artificial intelligence and smart home" which is a very weak objective.

Reviewer suggests to rewrite the paper with appropriate objective and outcome. Please refer to good survey papers.

Thank you

Author Response

Dear Reviewer,

We are truly grateful to yours critical comments and thoughtful suggestions. Based on these comments and suggestions, we have made careful modifications on the original manuscript. We hope the new manuscript will meet your standard. Below you will find our point-by-point responses to the comments:

Point 1: Their and too many grammatical errors starting from the abstract. Missing punctuations and other Typos is degrading the readability of the manuscript. Please revise the manuscript with appropriate proof reading.

Response 1: We already proofread our article and made some revision.

Point 2: Reviewer is unable to understand the novelty of the work. The outcome of the survey is specified as "our results suggest the relationship between artificial intelligence and smart home" which is a very weak objective. Reviewer suggests to rewrite the paper with appropriate objective and outcome.

Response 2: We revised our paper and add our contribution in the introduction section on page 3: “This research contributed to find out the trends of smart home technology and product and state the relationship between literature and product in this field. We conducted this review from the perspective of architecture and human concern.”

We also provide a criticism of the current state of the art of AI in smart homes on page 11: “Generally, there is room for further improvements of AI in smart home. Currently, smart home is more utilized in energy management, intelligent interaction and security with AI technologies of voice recognition and image recognition. In the foreseeable future, more and more products will use activity recognition, data processing and predictions making.”

Author Response File: Author Response.docx

Reviewer 4 Report

This paper presents an interesting literature review of the relation between smart building and artificial intelligence.

The significance of this paper should clearly outlined. Authors should provide more arguments about the "why of this analysis" as well as its expected impact:

The abstract should be improved concerning the question of Why ?

The conclusion should also be improved concerning the impact of this study.

Figure 6 presents an interesting synthesis. Could the authors present an explanation about the results of this figure.

The citation of references does not respect the journal requirements. It should be modfied.

Author Response

Dear Reviewer,

Point 1: The abstract should be improved concerning the question of Why ?

Response 1: We put an research question in abstract section on page 1: “In this paper, we aim to answer the research questions of ‘What is the trend of smart home technology and product’ and ‘what is the relationship between literature and product in smart home with AI’.”

Point 2: The conclusion should also be improved concerning the impact of this study.

Response 2: We rewrite the conclusion section and discuss it based on the result on page 11: “This study aims to reveal how AI makes home smart. To achieve the goal, many literature and products were reviewed. We found that AI technology helps smart home in device management, energy management, health care, intelligent interaction, security, entertainment system, and personal robot by utilizing activity recognition, data processing, decisions making, image recognition, predictions making, and voice recognition. There is a delay between the literature and the product, the product is concentrated in a relatively simple way like image recognition and voice recognition. Literature is concentrated in a relatively complicated way like activity recognition and predictions making. AI with voice and image recognition is widely used in smart home products while the technology of activity recognition, data processing, and predictions making still need to be developed.

Furthermore, an interesting finding in this study is that intelligent interaction becomes more and more important both in literature and product. In the foreseeable future, smart home will pay more attention to the interaction between people and the environment to make the building more sustainable and personalized. One important future direction to apply AI to smart home is considered at both smart home technology and architecture design and to develop relevant standards.”

Point 3: Figure 6 presents an interesting synthesis. Could the authors present an explanation about the results of this figure.

Response 3: We add more explanation about the results of Figure 6 on page 10: “These data are consistent with the notion in practice, AI is more often used in the identification and recognition of the primary stage, while activity recognition, data processing, decisions making, and prediction making require further development of artificial intelligence technology. From Figure 6, we can see the relationship between literature and products, that is, no one is in an absolute leading position. Literature is leading the way in complex technology of AI in recent years, while products are more subject to market. Therefore, once a technology is relatively mature, there will be more products using this technology.”

Point 4: The citation of references does not respect the journal requirements. It should be modfied.

Response 4: We’ve already revised the references according to MDPI Reference List and Citations Style Guide.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Thanks for addressing my comment. I believe the paper is good to go for publication. It is a decent paper for people that are new to the field or for people that are not engineers.

Author Response

Thank you very much for your kind comments on our manuscript.

Reviewer 2 Report

At the moment I still cannot recommend the paper towards acceptance, as the comments were not addressed in the due way.

The selected classification to conduct the research (the one with 6 categories/clusters, namely: activity recognition, data processing, decisions making, image recognition, predictions making, and voice recognition) appears to be chosen at random / in a not well justified way, and, as already pointed out in my first review, does not reflect adequately the field of AI. The categories seem to be derived from the work of Tang - reference 12 - however the reference does not match the number, and the reference 12 appears to be some other reference. So I could not see the paper, but I would not say that the survey coverage covers representatively the key fields of AI and smart home applications, as it promises. Particularly, these 6 categories do not cover the whole symbolic AI fields - or, in particularly, the completed survey - that are now very important, such as knowledge graphs, semantic technology, rule based systems. Mentioning these in one sentence does not solve the issue of the survey not correspondingly conducted.

The reviewed papers and pictures were also not updated accordingly (there are still the category names and papers from the previous version). So basically the requested major revision was not made.

An adequate well accepted AI fields/categories framework should be taken, and the survey (or, parts of it), should be updated/completed correspondingly.

There are still many issues with the language. The paper does not seem to be proofread by someone who is fluent enough/native in English. For example, it contains sentences with mistakes such as:

- "Seungmin et al. selected nine manuscripts of which the topics are strongly related to the
37 intelligent surveillance system in smart home environment to indicated that many researchers",

- "and the integration of literature and products were 137 not been fully
138 considered either." , and so on.

Also many sentences, though grammatically correct, are expressed in a very awkward manner.

Author Response

Thank you for your comments on our manuscript.

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

The paper still needs extensive English editing. For example, even the first line of the abstract is having grammatical errors.

Smart home and artificial intelligence technologies are developing rapidly, and various of 11 smart home products associated with artificial intelligence (AI) has improved the ability of 12 providing occupants with better comfort.

Reviewer suggests to have the paper corrected by English language expert

Author Response

Thank you for your comments on our manuscript. We proofread and revised it with the help of an English teacher who comes from America.

- Smart home and artificial intelligence technologies are developing rapidly, and various smart home products associated with artificial intelligence (AI) has improved the ability of the quality of living for occupants.