PONE-D-20-17716

A model based on Bayesian inference and machine learning algorithms to aid archaeological interpretation by integrating incompatible data

PLOS ONE

Dear Dr. Vos,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

Please submit your revised manuscript by Oct 29 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

• A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
• A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
• An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Ceren Kabukcu, PhD

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and

https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2.In your manuscript, please provide additional information regarding the specimens used in your study. Ensure that you have reported specimen numbers and complete repository information, including museum name and geographic location.

If permits were required, please ensure that you have provided details for all permits that were obtained, including the full name of the issuing authority, and add the following statement:

'All necessary permits were obtained for the described study, which complied with all relevant regulations.'

If no permits were required, please include the following statement:

'No permits were required for the described study, which complied with all relevant regulations.'

For more information on PLOS ONE's requirements for paleontology and archaeology research, see https://journals.plos.org/plosone/s/submission-guidelines#loc-paleontology-and-archaeology-research.

3.We note that [Figure(s) 1] in your submission contain [map/satellite] images which may be copyrighted. All PLOS content is published under the Creative Commons Attribution License (CC BY 4.0), which means that the manuscript, images, and Supporting Information files will be freely available online, and any third party is permitted to access, download, copy, distribute, and use these materials in any way, even commercially, with proper attribution. For these reasons, we cannot publish previously copyrighted maps or satellite images created using proprietary data, such as Google software (Google Maps, Street View, and Earth). For more information, see our copyright guidelines: http://journals.plos.org/plosone/s/licenses-and-copyright.

We require you to either (1) present written permission from the copyright holder to publish these figures specifically under the CC BY 4.0 license, or (2) remove the figures from your submission:

1.    You may seek permission from the original copyright holder of Figure(s) [1] to publish the content specifically under the CC BY 4.0 license. 

We recommend that you contact the original copyright holder with the Content Permission Form (http://journals.plos.org/plosone/s/file?id=7c09/content-permission-form.pdf) and the following text:

“I request permission for the open-access journal PLOS ONE to publish XXX under the Creative Commons Attribution License (CCAL) CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). Please be aware that this license allows unrestricted use and distribution, even commercially, by third parties. Please reply and provide explicit written permission to publish XXX under a CC BY license and complete the attached form.”

Please upload the completed Content Permission Form or other proof of granted permissions as an "Other" file with your submission.

In the figure caption of the copyrighted figure, please include the following text: “Reprinted from [ref] under a CC BY license, with permission from [name of publisher], original copyright [original copyright year].”

2.    If you are unable to obtain permission from the original copyright holder to publish these figures under the CC BY 4.0 license or if the copyright holder’s requirements are incompatible with the CC BY 4.0 license, please either i) remove the figure or ii) supply a replacement figure that complies with the CC BY 4.0 license. Please check copyright information on all replacement figures and update the figure caption with source information. If applicable, please specify in the figure caption text when a figure is similar but not identical to the original image and is therefore for illustrative purposes only.

The following resources for replacing copyrighted map figures may be helpful:

USGS National Map Viewer (public domain): http://viewer.nationalmap.gov/viewer/

The Gateway to Astronaut Photography of Earth (public domain): http://eol.jsc.nasa.gov/sseop/clickmap/

Maps at the CIA (public domain): https://www.cia.gov/library/publications/the-world-factbook/index.html and https://www.cia.gov/library/publications/cia-maps-publications/index.html

NASA Earth Observatory (public domain): http://earthobservatory.nasa.gov/

Landsat: http://landsat.visibleearth.nasa.gov/

USGS EROS (Earth Resources Observatory and Science (EROS) Center) (public domain): http://eros.usgs.gov/#

Natural Earth (public domain): http://www.naturalearthdata.com/

Additional Editor Comments (if provided):

Dear authors,

Many apologies for a lengthy peer-review and decision process. The reviews, as you will see below, are polarised. Reviewer 1 has suggested an outright rejection of the manuscript, while Reviewer 2 suggests Minor Revision. Having read through both sets of comments, the manuscript and made an evaluation, I am recommending major improvements to the manuscript in order to proceed with further peer review. In your resubmission, please address all points raised by both reviewers.

As highlighted by Reviewer 1, there are serious concerns with a thorough description and theoretical/methodological backing with regard to the use of Bayesian statistics in addressing the research questions raised in the manuscript. This point is also raised by Reviewer 2, albeit with different wording. Namely, the issue of identifying and determining the contextual interpretation of specific archaeological features and/or occupation areas is presented in the manuscript as being better represented through geochemical & phytolith analysis supported by statistical examination of the resulting datasets. However, the question still remains as to which method (i.e., laboratory analysis of materials deriving from a site or contextual observations during excavation) are or should be considered more definitive. As suggested by Reviewer 2, this theoretical point must be clarified throughout the text, making explicit the fact that in order to justify the interpretations presented in the manuscript further work at archaeological sites with more definitive contextual observations could follow the presented research in order to be able to test some of the assumptions.

With regard to the details on statistical methodologies used, please address all points raised by reviewer 1:

“From the theory presented in the paper (line 269), it seems that Bayes theorem may have been used to combine conditional probabilities. However, it is unclear why this specific equation is used. They do justify the prior probability, albeit later in the paper (lines 414 - 415). However, it is not clear how the rest of the equation on line 269 is derived. For a full inference approach, we would expect to see a clear statement of a likelihood or at least a series of conditional probability statements, but these are not present. Without formality of this sort, use of the words “Bayesian inference” and “model” in the title and text (lines 81, 83, 86, 89, 122, 264, 334, 338) are misleading.”

In light of these comments, I would suggest either a full-scale re-writing of the methodological aspects of the article, removing statements pertaining to Bayesian inference and modelling. Alternatively, you could provide a full and detailed rebuttal of the points raised by Reviewer 1, or could seek to modify your application of this aspect of the manuscript.

With regard to the application of PCA & multivariate analyses, please address all points raised by both reviewers. More specifically, please also provide further examples from already published studies as highlighted by Reviewer 2 (i.e., Rondelli et al., Lancelloti et al. and Peto et al.). With regard to the discussion of PCA and presentation of the results, I encourage the authors to also include a fuller justification of the choice of multivariate technique to be used on the datasets (see Reviewer 1 comments). For example, other methods such as Multiple Factor Analysis or other multivariate techniques enabling the analysis of datasets of mixed sources/scales of recording and/or different natures were not employed in the present study. Alternatively, if the aim of the study is to examine a characterisation/discrimination of context types with regard to the data categories included, other techniques such as cluster analysis or discriminant functions could have also been employed. Furthermore, I suggest that the authors also present a more solid justification with regard to the choice of 3D plotting, and provide the results of PCA in full as supplementary materials (e.g., contributions to dimensions, inertia etc).

As a final comment, while the text reads clear and concise, I ask that the authors revise certain statements throughout the text which refer to the nature of archaeological data as ambiguous or carrying an inherent ambiguity or vagueness (particularly with regard to contextual data collection during excavation). There is no need to view archaeological datasets, however fragmentary or incomplete, as any less rigorous and/or representative.

Kind Regards,

Ceren Kabukcu

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: No

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: No

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors of this paper suggest, in the title and abstract, that they are offering a novel, Bayesian, model-based approach for combining partial or incomplete archaeological data, thus improving the quality of the archaeological interpretation from a given site. However, what is presented in the main body of the paper is a collection of ad hoc tools with limited statistical or methodological motivation or justification. Here we outline four main concerns: lack of methodological motivation for the choice of methods; lack of discussion of other methods that could have been used; uncertainty as to whether Bayesian inference has actually been carried out; and weak motivation for and details of the statistical theory that is used.

The authors offer archaeological motivation of the need for methods to combine formally information from different sources, but they do not justify their selection of the particular suite of methods chosen and offer no evidence as to what other methods were considered. The authors do present some reasoning for using Bayesian inference (lines 121-125). However, this does not justify why using Bayesian inference in conjunction with classification trees is the most suitable method for combining related sources of rather limited archaeological data, of the sort described.

Furthermore, it is unclear from the text whether Bayesian inference has actually been carried out. From the theory presented in the paper (line 269), it seems that Bayes theorem may have been used to combine conditional probabilities. However, it is unclear why this specific equation is used. They do justify the prior probability, albeit later in the paper (lines 414 - 415). However, it is not clear how the rest of the equation on line 269 is derived. For a full inference approach, we would expect to see a clear statement of a likelihood or at least a series of conditional probability statements, but these are not present. Without formality of this sort, use of the words “Bayesian inference” and “model” in the title and text (lines 81, 83, 86, 89, 122, 264, 334, 338) are misleading.

Aside from the ambiguity regarding the use of Bayesian inference, it is also unclear why PCA is used in this paper. One might say that it was included as an exploratory data analysis tool, to give the reader a sense of the data. However, this has not been motivated and so its appearance (at line 158) is rather abrupt. Furthermore, later the authors appear initially to be using PCA to justify the success of their method (by showing improved clustering), which is misleading since they use the same data for both the classification trees and the PCA. The authors do address this multiple use of the same data in the subsequent paragraph (lines 363-369), but including this plot nonetheless suggests some circularity in the author's thinking.

Specific comments that we hope might help in any revisions to the text:

Throughout: Abbreviations such as PCA and XRF are not defined.

58: It would be helpful to reference some standard textbooks for readers unfamiliar with phtolyith analysis, micromorphology, or lipid residue analysis.

65: It is unclear what is meant by the phrase “higher interpretive power”. Since power has a technical statistical meaning, this may mislead the reader.

90: Not sure about the use of the word vagueness, what does this mean in the context of the data?

257-261: The meaning of this section is very unclear, but is vital to the value or otherwise of the work described. In order to fully understand and verify the methodology, considerably greater detail is needed here about the Weka software, the J48 classifier and the concepts of correctly and incorrectly classifying “according to set parameters”. Adoption of these tools also lacks any real motivation. Why these particular tools and not the many other similar ones available?

293-309: There seems to be a typo in one or other of these captions since they both refer to the same dataset (WJ13).

333-349 (and Table 3): The reviewers could not discern the methodology from the description offered here and, as a result, could not understand the contents of Table 3. Precisely what are the numbers in Table 3, how should we interpret them given the equation provided and why are so many of the numbers in each column identical?

Reviewer #2: The submitted manuscript presents a study on data integration within household archaeological research. The subject of the study is up-to-date, because the integration of data that have dissimilar dimensions is always an important and sometimes problematic part of multi-proxy approaches.

Lines 63-75: The circumstances of multi-proxy studies in general, which are described in this paragraph are true, however I would suggest to rephrase it a little, since there are existing ways - even if partly debatable - to perform data integration by the means of standardization and normalization. These preparatory tools in statistics can be utilized on data sets with different dimensions in order to integrate them within an interpretive multi elemental statistical approach such as PCA, Cluster and many others.

Attempts on this can be consulted in Rondelli et al. 2014 Anthropic activity markers and spatial variability: an ethnoarchaeological experiment in a domestic unit of northern Gujarat (India). J Archaeol Sci 41:482-492, or Lancelotti et al. Investigating fuel and fireplaces with a combination of phytoliths and multielement analysis; an ethnographic experiment VHA 26: 75-83, or Pető et al 2014: Activity area analysis of a Roman period semi-subterranean building by means of integrated archaeobotanical and geoarchaeological data VHA 24: 101-120.

In addition to the abovementioned this paper presents another and in this sense novel approach to data integration. It uses data from a previous study. Materials and methods, as well as data are partly presented and described in the introduction of the manuscript. The way the older data is presented is sufficient for the understanding of the new integration approach, however one must delve deep into the PhD dissertation of the corresponding author to get a more detailed picture.

Lines 152-153: Why is it interesting that K and Mg elevations occur in hearth areas? To my best knowledge ash contains a lot of these elements.

Lines 432-437: I think this is an evident statement throughout this profession therefore it doesn’t really need to be addressed here. This derives from the nature of multi-proxy studies that different elements of the ’whole mosaic’ can be reconstructed with the different pairing of the various approaches.

The main result of the paper is shown in Fig 13. The entire attempt focuses on the re-classification of the household contexts based on the integration of the phytolith and geochemical dataset. This is very important; however, the results do not appear plausible on this figure. It would be worth improving the visualization of this, because this is one of the key elements of the paper.

General comment: The paper attempts to show that the statistical way that was utilized is a proper way to aid the functional identification of archaeological features and contexts. The paper states that the case study was carried out on a site where the remains of the material culture are extremely sparse, therefore the identification of the archaeological features and contexts was revisited and certain modification could have been established. The concept of the paper is clear and the approach is fine, however how can the authors be sure that the re-classification of the features and contexts is valid if there is no archaeological evidence to support that? With other words, it would be great to repeat this study at a site which provides enough ‘classic’ archaeological evidence, so the blind test of the phytolith and geochemical data could be backed up and the method could be validated. This logic should be at least mentioned in the discussion so that readers understand the limitation of the approach.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

A model based on Bayesian confirmation and machine learning algorithms to aid archaeological interpretation by integrating incompatible data

PONE-D-20-17716R1

Dear Dr. Vos,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Peter F. Biehl, PhD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #3: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #3: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #3: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #3: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #3: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #3: This is a well-written and focused paper that is very easy to follow. The authors present their topic and case study in a natural and straightforward way, and are very conscientious about using unassuming language and explicitly introducing and defining concepts. As the authors suggest, this is important for addressing a general audience; however, I would go a step further and say it's a good practice in addressing ANY audience.

So far as I can tell, the statistical methods are sound. The assumptions in the model are fairly simplistic, but it doesn't have to be complex. If anything, I appreciate that it provides a clear illustration of the methods used without the burden of a highly complex set of assumptions or a huge mass of data. The data used are all provided and steps are clearly outlined as to how to replicate the analysis.

The concerns of previous reviewers seem to have been adequately addressed.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #3: No