Conditional Dependencies and Position Fixing

Round 1

Reviewer 1 Report

Dear Authors,

You need to improve the manuscript to bring it to the level of scientific writing style. It is very hard for another person to understand your research.

1 – The title of the manuscript is not representing the research. Please fix and modify the title.

2 – Abstract should clearly describe the research problem, what is the proposed research method to solve the problem, and what is the result. In your abstract result is missing completely.

3 – The introduction needs modification. Need to add more prior and related work. You have insufficient references. You must increase the references.

4 – Please write the exact research contribution after the introduction and define the problem.

4 – Figure 1, 2, 3, 4, 7: Axis title, values and labels are missing.

5 – Figure 5, 6: Axis title and labels are missing

6 – The result presented in Table 2 should be explained more clearly.

Author Response

Title:

My intention was presenting position fixing as an inference scheme. In nautical science it is novelty thus bibliography is rather scarce. The scheme engages evidence, hypothesis and revokes concept of conditional relationships. In this view, title “conditional dependencies and position fixing” is relevant to my objective.

Abstract:

Experimental crude data sets are usually processed with statistic methods in order to get rough evaluations of nautical measurements. Taking observations and rectifying knowledge on them are not correlated. In modern computer applications, raw data sets are usually exploited at initial, learning phase. At this stage, available data are explored in order to extract necessary parameters required within the scheme of computations. The paper undertakes the crude data processing problem in order to extract conditional dependencies that appears as the most important factor while handling distorted data. Traditional structures, histograms are upgraded at first. Step-wise diagrams feature direct ability of their uncertainty evaluation. Hierarchy among evidence within the pool of data is upgraded. For given ranking adequate membership functions are defined. Principles of fuzzy systems justify using bin-to-bin additive approach in order to obtain locally injective density functions. They can be perceived as conditional dependency diagrams enabling constructing simple belief assignments. The structures combination include solution to position fixing problem.

Introduction

Dealing with imprecise data requires models that include knowledge into the framework. To meet this requirement one should engage the fuzzy systems and Mathematical Theory of Evidence (MTE for short), which deliver platform for processing uncertainty [1, 2, 3]. Models require methods to obtain formal evaluation of uncertainty. In nautical science, the knowledge refers to distribution of random variables instances. Usually they are perceived as governed by Gaussian dispersion patterns. Knowing the magnitude of standard deviations enables introduction of an observation rough assessment in terms of good, bad or sufficient. This attitude is popular among specialists. For those who were involved in implementing computer procedures variety of a measurement quality are perceived as uncertainty, and should be included it into processing scheme. The concept was discussed in the author’s publications, [4] is an example, proposed model accepted nautical knowledge thanks to evidence proximity exploration and engaging principles of fuzzy systems. Based on available navigation systems standard deviations adequate membership functions were defined and input into the processing. In proposed approach, it is an application where analyses of available raw recorded instances are carried out in order to extract even wider range of useful parameters. Their scope and methods of their evaluation are proposed. Thanks to this, prior estimations are of secondary meaning and do not directly affect the final solutions. Note that the former evaluations might be unavailable thus suggested approach appears as universal one whenever random data are involved. Let us also emphasis that evidence engaged in the problem should be normalized. Expected is ranking regarding decisiveness on affecting the solution. Items with lower uncertainty are of greatest influence in this respect. The paper contains section devoted to defining hierarchy among evidence within a pool of data.

line 390.

Table 2 contains the conditional dependencies calculated base on those read from the diagrams. For every hypothesis point, belief assignments using formula 8 were obtained referring each piece of evidence. Conjunctive combination of all structures result in cumulative measures. Final plausibility on representing a fixed position by each of hypotheses point presents last column of the table.

Conclusions

The new approach towards imprecise data handling was proposed. Inference computation scheme were followed while solving nautical position fixed problem. Raw experimental instances, available with statistically sufficient count, were assumed as initial source of knowledge regarding behaviour of available systems indications. Modern applications include learning phase, while necessary elements are extracted for further usage. In proposed approach conditional dependencies remain to be discovered. Appropriate diagrams take forms of converted histograms. Their construction took place during one of the first processing steps. Evaluation of the constructs reliability is an important issue. Uncertainty regarding vertical and horizontal layout of a histogram were considered. Measures referred to variety of bin heights and to their widths enabled estimation of an overall uncertainty. Doubtfulness and shape of the membership functions are mutually dependent. Both were main factor that decide on shape of converted histograms. They can be seen as conditional dependencies adequate diagrams. Pool of data required for solving problem engaging distorted data, making a fix is an example, needs additional processing. Carried out normalization introduces uniform probability density distributions. Relative initial vector is constructed in order to obtain the required hierarchy and subsequently probability assignments. Their combination deliver solution of the problem.

Proposed scheme of solving problems involving distorted data includes several steps:

1. record statistically justified number of instances for involved sources of random data;
2. explore stored sets in order to mine embedded uncertainty. Suggested solution involves histograms, the way of their evaluation was presented in the paper;
3. create ranking regarding included doubtfulness within the available evidence items;
4. given uncertainty depended membership functions and bin-to-bin additive method obtain locally injective density distribution diagrams consider them as conditional dependencies;
5. use the diagrams and upgrade MTE’s simple belief assignments. Result of their associations include measures that indicate the solution.

Reviewer 2 Report

The novelty of the proposal does not present the level required by the journal

 

Author Response

no comment

Reviewer 3 Report

This paper undertakes the crude data processing problem in order to extract conditional dependencies that appears as the most important factor while handling distorted data.

The work has been prepared in an innovative way. However abstract and introduction sections need to improved. Some latest references related to research should be added in the reference section. 

"figure 1" will be replaced by "Figure 1" in description. Similarly for other figures.

 

Author Response

Title:

My intention was presenting position fixing as an inference scheme. In nautical science it is novelty thus bibliography is rather scarce. The scheme engages evidence, hypothesis and revokes concept of conditional relationships. In this view, title “conditional dependencies and position fixing” is relevant to my objective.

Abstract:

Experimental crude data sets are usually processed with statistic methods in order to get rough evaluations of nautical measurements. Taking observations and rectifying knowledge on them are not correlated. In modern computer applications, raw data sets are usually exploited at initial, learning phase. At this stage, available data are explored in order to extract necessary parameters required within the scheme of computations. The paper undertakes the crude data processing problem in order to extract conditional dependencies that appears as the most important factor while handling distorted data. Traditional structures, histograms are upgraded at first. Step-wise diagrams feature direct ability of their uncertainty evaluation. Hierarchy among evidence within the pool of data is upgraded. For given ranking adequate membership functions are defined. Principles of fuzzy systems justify using bin-to-bin additive approach in order to obtain locally injective density functions. They can be perceived as conditional dependency diagrams enabling constructing simple belief assignments. The structures combination include solution to position fixing problem.

Introduction

Dealing with imprecise data requires models that include knowledge into the framework. To meet this requirement one should engage the fuzzy systems and Mathematical Theory of Evidence (MTE for short), which deliver platform for processing uncertainty [1, 2, 3]. Models require methods to obtain formal evaluation of uncertainty. In nautical science, the knowledge refers to distribution of random variables instances. Usually they are perceived as governed by Gaussian dispersion patterns. Knowing the magnitude of standard deviations enables introduction of an observation rough assessment in terms of good, bad or sufficient. This attitude is popular among specialists. For those who were involved in implementing computer procedures variety of a measurement quality are perceived as uncertainty, and should be included it into processing scheme. The concept was discussed in the author’s publications, [4] is an example, proposed model accepted nautical knowledge thanks to evidence proximity exploration and engaging principles of fuzzy systems. Based on available navigation systems standard deviations adequate membership functions were defined and input into the processing. In proposed approach, it is an application where analyses of available raw recorded instances are carried out in order to extract even wider range of useful parameters. Their scope and methods of their evaluation are proposed. Thanks to this, prior estimations are of secondary meaning and do not directly affect the final solutions. Note that the former evaluations might be unavailable thus suggested approach appears as universal one whenever random data are involved. Let us also emphasis that evidence engaged in the problem should be normalized. Expected is ranking regarding decisiveness on affecting the solution. Items with lower uncertainty are of greatest influence in this respect. The paper contains section devoted to defining hierarchy among evidence within a pool of data.

line 390.

Table 2 contains the conditional dependencies calculated base on those read from the diagrams. For every hypothesis point, belief assignments using formula 8 were obtained referring each piece of evidence. Conjunctive combination of all structures result in cumulative measures. Final plausibility on representing a fixed position by each of hypotheses point presents last column of the table.

Conclusions

The new approach towards imprecise data handling was proposed. Inference computation scheme were followed while solving nautical position fixed problem. Raw experimental instances, available with statistically sufficient count, were assumed as initial source of knowledge regarding behaviour of available systems indications. Modern applications include learning phase, while necessary elements are extracted for further usage. In proposed approach conditional dependencies remain to be discovered. Appropriate diagrams take forms of converted histograms. Their construction took place during one of the first processing steps. Evaluation of the constructs reliability is an important issue. Uncertainty regarding vertical and horizontal layout of a histogram were considered. Measures referred to variety of bin heights and to their widths enabled estimation of an overall uncertainty. Doubtfulness and shape of the membership functions are mutually dependent. Both were main factor that decide on shape of converted histograms. They can be seen as conditional dependencies adequate diagrams. Pool of data required for solving problem engaging distorted data, making a fix is an example, needs additional processing. Carried out normalization introduces uniform probability density distributions. Relative initial vector is constructed in order to obtain the required hierarchy and subsequently probability assignments. Their combination deliver solution of the problem.

Proposed scheme of solving problems involving distorted data includes several steps:

1. record statistically justified number of instances for involved sources of random data;
2. explore stored sets in order to mine embedded uncertainty. Suggested solution involves histograms, the way of their evaluation was presented in the paper;
3. create ranking regarding included doubtfulness within the available evidence items;
4. given uncertainty depended membership functions and bin-to-bin additive method obtain locally injective density distribution diagrams consider them as conditional dependencies;
5. use the diagrams and upgrade MTE’s simple belief assignments. Result of their associations include measures that indicate the solution.

Reviewer 4 Report

The paper can be accepted for publication if the authors revise the paper to be clearer and understandable. As summarized in the following:

• The paper's literature positioning in Conditional dependencies and position fixing is unclear. It is important to update the list of references with respect to the latest published papers in mainstream leading journals in the science and technology of sensor and its applications and management. This will highlight the paper's novelty and literature positioning in the area.
•  Motivation and model setup: Proper, convincing and timely international motivation cases should be included.
• The English of the paper is not clear in several parts, and some parts are not clear enough to understand the authors' idea. The English should be improved, and the grammatical mistakes should be corrected. The abstract should be rewritten carefully and the novelty of the work should be explained clearly in the abstract.

• Generally speaking, the introduction section should be improved. The problem and the research question (s) are not clearly stated. Also, the description of the value proposition of this study must be revised.
• A clearer definition and description of the new contribution to the work should be presented at the end of the introduction section. Please add a research gap section.

• The overall methodology of the present study should be clarified to be more understandable.

• What could be the limitation of the proposed method/analysis procedure? may be further explained.

·         The managerial implications of this study, i.e. "what's the big deal?", are not well-explained. How would Feedback Integrators be benefited from the findings of your study? What are the specific action plans based on the research findings? These should be addressed.

·         Please highlight the contribution of this study, with reference to practice and contribution to the literature and solution methodology. In particular, some advances in solution methodology should be developed for optimization based research.

·         It is critical to compare the findings with the prior literature as well as check whether the findings can challenge/support the current industrial practice (i.e., industrial verification).

·         How robust are the findings?

·         A table should be added to show a summary of the simulated and numerical results. This will make the results easier to understand.

• The conclusion should be explained technically, and it needs to be rewritten.

Author Response

Title:

My intention was presenting position fixing as an inference scheme. In nautical science it is novelty thus bibliography is rather scarce. The scheme engages evidence, hypothesis and revokes concept of conditional relationships. In this view, title “conditional dependencies and position fixing” is relevant to my objective.

Abstract:

Experimental crude data sets are usually processed with statistic methods in order to get rough evaluations of nautical measurements. Taking observations and rectifying knowledge on them are not correlated. In modern computer applications, raw data sets are usually exploited at initial, learning phase. At this stage, available data are explored in order to extract necessary parameters required within the scheme of computations. The paper undertakes the crude data processing problem in order to extract conditional dependencies that appears as the most important factor while handling distorted data. Traditional structures, histograms are upgraded at first. Step-wise diagrams feature direct ability of their uncertainty evaluation. Hierarchy among evidence within the pool of data is upgraded. For given ranking adequate membership functions are defined. Principles of fuzzy systems justify using bin-to-bin additive approach in order to obtain locally injective density functions. They can be perceived as conditional dependency diagrams enabling constructing simple belief assignments. The structures combination include solution to position fixing problem.

Introduction

Dealing with imprecise data requires models that include knowledge into the framework. To meet this requirement one should engage the fuzzy systems and Mathematical Theory of Evidence (MTE for short), which deliver platform for processing uncertainty [1, 2, 3]. Models require methods to obtain formal evaluation of uncertainty. In nautical science, the knowledge refers to distribution of random variables instances. Usually they are perceived as governed by Gaussian dispersion patterns. Knowing the magnitude of standard deviations enables introduction of an observation rough assessment in terms of good, bad or sufficient. This attitude is popular among specialists. For those who were involved in implementing computer procedures variety of a measurement quality are perceived as uncertainty, and should be included it into processing scheme. The concept was discussed in the author’s publications, [4] is an example, proposed model accepted nautical knowledge thanks to evidence proximity exploration and engaging principles of fuzzy systems. Based on available navigation systems standard deviations adequate membership functions were defined and input into the processing. In proposed approach, it is an application where analyses of available raw recorded instances are carried out in order to extract even wider range of useful parameters. Their scope and methods of their evaluation are proposed. Thanks to this, prior estimations are of secondary meaning and do not directly affect the final solutions. Note that the former evaluations might be unavailable thus suggested approach appears as universal one whenever random data are involved. Let us also emphasis that evidence engaged in the problem should be normalized. Expected is ranking regarding decisiveness on affecting the solution. Items with lower uncertainty are of greatest influence in this respect. The paper contains section devoted to defining hierarchy among evidence within a pool of data.

line 390.

Table 2 contains the conditional dependencies calculated base on those read from the diagrams. For every hypothesis point, belief assignments using formula 8 were obtained referring each piece of evidence. Conjunctive combination of all structures result in cumulative measures. Final plausibility on representing a fixed position by each of hypotheses point presents last column of the table.

Conclusions

The new approach towards imprecise data handling was proposed. Inference computation scheme were followed while solving nautical position fixed problem. Raw experimental instances, available with statistically sufficient count, were assumed as initial source of knowledge regarding behaviour of available systems indications. Modern applications include learning phase, while necessary elements are extracted for further usage. In proposed approach conditional dependencies remain to be discovered. Appropriate diagrams take forms of converted histograms. Their construction took place during one of the first processing steps. Evaluation of the constructs reliability is an important issue. Uncertainty regarding vertical and horizontal layout of a histogram were considered. Measures referred to variety of bin heights and to their widths enabled estimation of an overall uncertainty. Doubtfulness and shape of the membership functions are mutually dependent. Both were main factor that decide on shape of converted histograms. They can be seen as conditional dependencies adequate diagrams. Pool of data required for solving problem engaging distorted data, making a fix is an example, needs additional processing. Carried out normalization introduces uniform probability density distributions. Relative initial vector is constructed in order to obtain the required hierarchy and subsequently probability assignments. Their combination deliver solution of the problem.

Proposed scheme of solving problems involving distorted data includes several steps:

1. record statistically justified number of instances for involved sources of random data;
2. explore stored sets in order to mine embedded uncertainty. Suggested solution involves histograms, the way of their evaluation was presented in the paper;
3. create ranking regarding included doubtfulness within the available evidence items;
4. given uncertainty depended membership functions and bin-to-bin additive method obtain locally injective density distribution diagrams consider them as conditional dependencies;
5. use the diagrams and upgrade MTE’s simple belief assignments. Result of their associations include measures that indicate the solution.

Reviewer 5 Report

This paper undertakes the crude data processing problem in order to extract conditional dependencies that appears as the most important factor while handling distorted data. There are several questions requiring the author clarfify further in the revised version before the acceptance of the article.

1.  In Abstract, the description about the background of the problem proposed seems to be too long while the description for the work the author has done seems to be short. Therefore, the reviewer suggests the author moderately shorten the background description and strengthen the work description.

2. On the contrary, in Introduction there is only one paragraph which is insufficient to give the readers the whole study background. At the same time, the number of cited papers seem to be too less. Therefore, it is suggested that the author  strengthen the Introduction by adding more papers concerning this topic and describing the existed works by others.

3.  The Conclusions should be given point by point, but not a whole paragraph. Thus the potential readers may grasp the main concluding remarks obtained in this article.

4. Other small typos should be rectified, for example, at lines 263, 266 and 305.

Author Response

Title:

My intention was presenting position fixing as an inference scheme. In nautical science it is novelty thus bibliography is rather scarce. The scheme engages evidence, hypothesis and revokes concept of conditional relationships. In this view, title “conditional dependencies and position fixing” is relevant to my objective.

Abstract:

Experimental crude data sets are usually processed with statistic methods in order to get rough evaluations of nautical measurements. Taking observations and rectifying knowledge on them are not correlated. In modern computer applications, raw data sets are usually exploited at initial, learning phase. At this stage, available data are explored in order to extract necessary parameters required within the scheme of computations. The paper undertakes the crude data processing problem in order to extract conditional dependencies that appears as the most important factor while handling distorted data. Traditional structures, histograms are upgraded at first. Step-wise diagrams feature direct ability of their uncertainty evaluation. Hierarchy among evidence within the pool of data is upgraded. For given ranking adequate membership functions are defined. Principles of fuzzy systems justify using bin-to-bin additive approach in order to obtain locally injective density functions. They can be perceived as conditional dependency diagrams enabling constructing simple belief assignments. The structures combination include solution to position fixing problem.

Introduction

Dealing with imprecise data requires models that include knowledge into the framework. To meet this requirement one should engage the fuzzy systems and Mathematical Theory of Evidence (MTE for short), which deliver platform for processing uncertainty [1, 2, 3]. Models require methods to obtain formal evaluation of uncertainty. In nautical science, the knowledge refers to distribution of random variables instances. Usually they are perceived as governed by Gaussian dispersion patterns. Knowing the magnitude of standard deviations enables introduction of an observation rough assessment in terms of good, bad or sufficient. This attitude is popular among specialists. For those who were involved in implementing computer procedures variety of a measurement quality are perceived as uncertainty, and should be included it into processing scheme. The concept was discussed in the author’s publications, [4] is an example, proposed model accepted nautical knowledge thanks to evidence proximity exploration and engaging principles of fuzzy systems. Based on available navigation systems standard deviations adequate membership functions were defined and input into the processing. In proposed approach, it is an application where analyses of available raw recorded instances are carried out in order to extract even wider range of useful parameters. Their scope and methods of their evaluation are proposed. Thanks to this, prior estimations are of secondary meaning and do not directly affect the final solutions. Note that the former evaluations might be unavailable thus suggested approach appears as universal one whenever random data are involved. Let us also emphasis that evidence engaged in the problem should be normalized. Expected is ranking regarding decisiveness on affecting the solution. Items with lower uncertainty are of greatest influence in this respect. The paper contains section devoted to defining hierarchy among evidence within a pool of data.

line 390.

Table 2 contains the conditional dependencies calculated base on those read from the diagrams. For every hypothesis point, belief assignments using formula 8 were obtained referring each piece of evidence. Conjunctive combination of all structures result in cumulative measures. Final plausibility on representing a fixed position by each of hypotheses point presents last column of the table.

Conclusions

The new approach towards imprecise data handling was proposed. Inference computation scheme were followed while solving nautical position fixed problem. Raw experimental instances, available with statistically sufficient count, were assumed as initial source of knowledge regarding behaviour of available systems indications. Modern applications include learning phase, while necessary elements are extracted for further usage. In proposed approach conditional dependencies remain to be discovered. Appropriate diagrams take forms of converted histograms. Their construction took place during one of the first processing steps. Evaluation of the constructs reliability is an important issue. Uncertainty regarding vertical and horizontal layout of a histogram were considered. Measures referred to variety of bin heights and to their widths enabled estimation of an overall uncertainty. Doubtfulness and shape of the membership functions are mutually dependent. Both were main factor that decide on shape of converted histograms. They can be seen as conditional dependencies adequate diagrams. Pool of data required for solving problem engaging distorted data, making a fix is an example, needs additional processing. Carried out normalization introduces uniform probability density distributions. Relative initial vector is constructed in order to obtain the required hierarchy and subsequently probability assignments. Their combination deliver solution of the problem.

Proposed scheme of solving problems involving distorted data includes several steps:

1. record statistically justified number of instances for involved sources of random data;
2. explore stored sets in order to mine embedded uncertainty. Suggested solution involves histograms, the way of their evaluation was presented in the paper;
3. create ranking regarding included doubtfulness within the available evidence items;
4. given uncertainty depended membership functions and bin-to-bin additive method obtain locally injective density distribution diagrams consider them as conditional dependencies;
5. use the diagrams and upgrade MTE’s simple belief assignments. Result of their associations include measures that indicate the solution.

Round 2

Reviewer 1 Report

Dear Authors,

 

Thank you for addressing the comments.

 

Good Luck!!!

Author Response

Thank you

Reviewer 4 Report

The authors revise the paper to be clearer and understandable. As summarized in the following:

- The paper's literature positioning in Conditional dependencies and position fixing can be more improved.

- The English should be improved, and the grammatical mistakes should be corrected.

- What could be the limitation of the proposed method/analysis procedure? may be further explained.

Author Response

Conditional dependencies and position fixing problem literature is rather scarce. Papers delivered by the author are represented by the most significant publications.

Sorry, but I am not a native speaker.

Limitations are to be discovered. So far, proposed approach was tested with object classification and position fixing problems. It behaved correctly. It was also proposed for metrology and nautical science as a new, alternative concept. Potential applications were suggested.

Thank you