Paper: 298637 Title: Neural Network Congestion Control Design for TCP/IP Network Based Genetic Algorithm -------------------- review 1 -------------------- ---------------------------- REVIEW 1 -------------------------- PAPER: 2 TITLE: Neural Network Congestion Control Design for TCP/IP Network Based Genetic Algorithm OVERALL RATING: 2 (accept) REVIEWER'S CONFIDENCE: 2 (medium) Relevance to this conference: 4 (good) Originality/Uniqueness: 4 (good) English readability: 3 (fair) Paper organization/presentation: 3 (fair) Has good survey been done?: 3 (fair) COMMENTS: 1. On page 3, First equation should be PI, should not it. 2. The authors describe the use of a dynamic recurrent neural model. However, Fig. 5 shows only a feedforward network. I wonder where the recurrent neural network involves in the controller. 3. Although, the authors describe the use of GA for changing weights of the network, it is still unclear how the weights of the network are really updated. For clarity, I would recommend to also present the equations of weight changes. -------------------- review 2 -------------------- ---------------------------- REVIEW 2 -------------------------- PAPER: 2 TITLE: Neural Network Congestion Control Design for TCP/IP Network Based Genetic Algorithm OVERALL RATING: 2 (accept) REVIEWER'S CONFIDENCE: 3 (high) Relevance to this conference: 4 (good) Originality/Uniqueness: 4 (good) English readability: 4 (good) Paper organization/presentation: 5 (excellent) Has good survey been done?: 4 (good) Authors introduce a novel AQM methodology to overcome problems that are identifying to TCP networks dynamics due. Methodology can improve the type of controller design to previous controller design (PI, RED, PID and APC). ********************************************* Paper: 320590 Title: Functional Learning & Introduction to the Electronic Brain on El-Dorra neural network tehnology -------------------- review 1 -------------------- ---------------------------- REVIEW 1 -------------------------- PAPER: 3 TITLE: Functional Learning & Introduction to the Electronic Brain on El-Dorra neural network tehnology OVERALL RATING: 1 (weak accept) REVIEWER'S CONFIDENCE: 2 (medium) Relevance to this conference: 3 (fair) Originality/Uniqueness: 3 (fair) English readability: 2 (poor) Paper organization/presentation: 2 (poor) Has good survey been done?: 2 (poor) The paper is rather a tutorial than research paper (it refers to XOR problem and there are no author's experiments in it). However, the idea of separete learning is interesting (since it is linked to the human brain) and should be popularized. So I would advice to publish the paper but after strong correction of the paper organisation. Detailed remarks: 1. There are no references to other researches of separate learning 2. I would advise to change the style and language of presentation to more scientific one. 3. The paper layout should be rearanged - there are very weak connections between succeeding parts of the paper and therefore the paper is very difficult to read. 4. I would introduce to Introduction the description of the whole paper layout - it would make the paper more readible. -------------------- review 2 -------------------- ---------------------------- REVIEW 2 -------------------------- PAPER: 3 TITLE: Functional Learning & Introduction to the Electronic Brain on El-Dorra neural network tehnology OVERALL RATING: -3 (strong reject) REVIEWER'S CONFIDENCE: 2 (medium) Relevance to this conference: 4 (good) Originality/Uniqueness: 3 (fair) English readability: 2 (poor) Paper organization/presentation: 2 (poor) Has good survey been done?: 2 (poor) It sounds unique when reviewer reads the description by the author that "we can change the activate functions and we fix the weights," but the author actually fails to appeal how it works better than the standard approach. However, it's not the reason for the bad score. What is very bad is, the author seems to present exactly the same topic, method, and conclusion repeatedly in the different conferences without mentioning in this submission paper. For example, in the 4th International Conference on Neural, Parallel & Scientific Computations. If the author survives the review procedure, he definitely should show, in his camera-ready version, how different this paper is from his previous reports, if any. -------------------- review 3 -------------------- ---------------------------- REVIEW 3 -------------------------- PAPER: 3 TITLE: Functional Learning & Introduction to the Electronic Brain on El-Dorra neural network tehnology OVERALL RATING: 1 (weak accept) REVIEWER'S CONFIDENCE: 2 (medium) Relevance to this conference: 5 (excellent) Originality/Uniqueness: 4 (good) English readability: 3 (fair) Paper organization/presentation: 3 (fair) Has good survey been done?: 4 (good) The paper describes a different learning method & an electronic brain model. More specifically, authors introduce Electronic Brain on El-dorra neural network technology. The paper is presenting a really novel and interesting technology. Technically and scientifically is a very good work. However, I believe there is still place for more extended evaluation and conclusions. Furthermore, the paper doesn�t follow the instructions for authors and English readability is rather fair. Concluding, I suggest to be accepted (with minor changes). -------------------- review 4 -------------------- ---------------------------- REVIEW 4 -------------------------- PAPER: 3 TITLE: Functional Learning & Introduction to the Electronic Brain on El-Dorra neural network tehnology OVERALL RATING: 1 (weak accept) REVIEWER'S CONFIDENCE: 2 (medium) Relevance to this conference: 4 (good) Originality/Uniqueness: 4 (good) English readability: 3 (fair) Paper organization/presentation: 2 (poor) Has good survey been done?: 3 (fair) Authors present a new learning method. The paper describes Electronic Brain on El-dorra neural network technology. The paper is a very good technical document.The proposed technology is presented in a very sophisticated way. However, literature survey is rather limited. Also I would like to see more discussion concerning comparison with similar techniques. I believe that in some cases it lacks in depth discussion. English readability is fair. Furthermore, authors should follow the conference template. I suggest to accept the paper (authors could follow previous comments). ********************************************* Paper: 340024 Title: Pattern recognition with adaptive morphological composite filters -------------------- review 1 -------------------- ---------------------------- REVIEW 1 -------------------------- PAPER: 4 TITLE: Pattern recognition with adaptive morphological composite filters OVERALL RATING: 1 (weak accept) REVIEWER'S CONFIDENCE: 3 (high) Relevance to this conference: 4 (good) Originality/Uniqueness: 3 (fair) English readability: 4 (good) Paper organization/presentation: 3 (fair) Has good survey been done?: 1 (very poor) The article presents an application of adaptive correlation filters used to remove the background noise to increase pattern recognition capabilities. It contains the results of experiments over object recognition by the background removal. The article misses the-state-of-the-art section. It contains few citations, but they do not address already existing applications of correlation filters only mention three kinds of filtering algorithms. The reference section if very limited, it should contain more positions, proving that a detailed survey was done. It would be easier for reader to understand section 2 if author provide a picture presenting the idea of a moving window (the first paragraph in this section). The equations should be centred. The figure 2 shows the objects to be recognised and background to be removed. In my opinion this section misses an example of noisy image and filter output i.e. pre-processed image. It would make easier to understand the example by seeing the images being filter's income and outcome. Moreover, it would be more simulative to show another test case so reader could evaluate filters comparing the results achieved for the two images. Currently it is questionable how the results were influenced by the selection of filters and how by the selection of the test image. -------------------- review 2 -------------------- ---------------------------- REVIEW 2 -------------------------- PAPER: 4 TITLE: Pattern recognition with adaptive morphological composite filters OVERALL RATING: 2 (accept) REVIEWER'S CONFIDENCE: 3 (high) Relevance to this conference: 3 (fair) Originality/Uniqueness: 3 (fair) English readability: 4 (good) Paper organization/presentation: 4 (good) Has good survey been done?: 4 (good) Authors propose and design adaptive morphological composite correlation filters for robust and distortion-invariant pattern recognition. The recognition performance of the proposed filters is evaluated and computer simulation results are provided and discussed. In general, the authors cover the overall subject in a critical way. The paper is well presented and organised. I believe that the paper is comprehensive, original and would be important to researchers within the pattern recognition field. Therefore, I suggest to be accepted. -------------------- review 3 -------------------- ---------------------------- REVIEW 3 -------------------------- PAPER: 4 TITLE: Pattern recognition with adaptive morphological composite filters OVERALL RATING: 2 (accept) REVIEWER'S CONFIDENCE: 3 (high) Relevance to this conference: 3 (fair) Originality/Uniqueness: 4 (good) English readability: 4 (good) Paper organization/presentation: 3 (fair) Has good survey been done?: 3 (fair) In this paper, the design of adaptive morphological composite correlation filters for robust pattern recognition is presented in detail. Furthermore, the authors evaluate the performance of the proposed filters in a very critical way. Computer simulation results are also presented in a detail way. I believe, that the presented work is original and novel. The paper is well organised. The proposed method would be very interesting to conferences attendees. Concluding, I suggest to be accepted as it is. ********************************************* Paper: 346758 Title: Models of Cognitive Evolution: Initial Steps -------------------- review 1 -------------------- ---------------------------- REVIEW 1 -------------------------- PAPER: 5 TITLE: Models of Cognitive Evolution: Initial Steps OVERALL RATING: 2 (accept) REVIEWER'S CONFIDENCE: 2 (medium) Relevance to this conference: 4 (good) Originality/Uniqueness: 3 (fair) English readability: 4 (good) Paper organization/presentation: 4 (good) Has good survey been done?: 3 (fair) The paper is relevant to the theme of the conference; it presents some approaches to modeling cognitive evolution, albeit on quite primitive tasks, far from practice. The article is written in good English, although sometimes there are typos (in abstract and probably couple of brackets are missed in the expression 2). The material is consistent and well organized, but in my opinion attention paid to consideration of similar approaches is not enough. -------------------- review 2 -------------------- ---------------------------- REVIEW 2 -------------------------- PAPER: 5 TITLE: Models of Cognitive Evolution: Initial Steps OVERALL RATING: 2 (accept) REVIEWER'S CONFIDENCE: 3 (high) Relevance to this conference: 4 (good) Originality/Uniqueness: 4 (good) English readability: 4 (good) Paper organization/presentation: 4 (good) Has good survey been done?: 4 (good) Authors describe the design of three initial models of autonomous agents that have elementary cognitive features. In general the paper is very well written and well structured. Authors cover the overall subject in an efficient and critical way. I believe that it would appeal to a wide audience of the conference, especially to researchers within the paper field. Therefore, I suggest to be accepted. ********************************************* Paper: 346955 Title: Kalman Gain Calculations with a Neural Network -------------------- review 1 -------------------- ---------------------------- REVIEW 1 -------------------------- PAPER: 6 TITLE: Kalman Gain Calculations with a Neural Network OVERALL RATING: 2 (accept) REVIEWER'S CONFIDENCE: 3 (high) Relevance to this conference: 4 (good) Originality/Uniqueness: 4 (good) English readability: 4 (good) Paper organization/presentation: 4 (good) Has good survey been done?: 4 (good) The paper is good. In my opinion, the principles of operation of a back-propagation neural network at matrix inversion should be described shortly. This operation is the key method of the paper, so it should be explained in some details. Also, results of simulations should be explained more clearly. Some technical recommendations are below. 1. References should be made in accordance with format requirements of the ICNNAI-2010. 2. Symbols in figures 1 and 2 are too small. It is difficult to read these notations. 3. Variables in the text should be made more accurately. Final conclusion: the paper can be accepted after small improvement. -------------------- review 2 -------------------- ---------------------------- REVIEW 2 -------------------------- PAPER: 6 TITLE: Kalman Gain Calculations with a Neural Network OVERALL RATING: 2 (accept) REVIEWER'S CONFIDENCE: 4 (expert) Relevance to this conference: 5 (excellent) Originality/Uniqueness: 4 (good) English readability: 4 (good) Paper organization/presentation: 4 (good) Has good survey been done?: 1 (very poor) The article presents an interesting way of gain ration calculation for the Kalman filter via neural network, eliminating a costly matrix inversion step influencing the stability of filters. The state of the art section is missing, thus it isn't clear if the proposed idea is an author's invention or he/she examine an already proposed solution still uncommon and hopes to popularise it. The revised paper should explain this issue. As the original Kalman filter is a bit out-dated (due to linear characteristic) and could be exchanged with modern nonlinear filters e.g. unscented or particle filters, a question arises - are there any future plans to use a similar approach to other filters e.g. UKF or CDKF? I hope the revised paper will explain this question. The article should follow the available draft - the reference section does not use Chicago citation style. I also recommend to point-out the three sentences following "Some of them are: " (page 1) and to centre all the equations. -------------------- review 3 -------------------- ---------------------------- REVIEW 3 -------------------------- PAPER: 6 TITLE: Kalman Gain Calculations with a Neural Network OVERALL RATING: 2 (accept) REVIEWER'S CONFIDENCE: 2 (medium) Relevance to this conference: 4 (good) Originality/Uniqueness: 3 (fair) English readability: 4 (good) Paper organization/presentation: 4 (good) Has good survey been done?: 3 (fair) The paper presents an approach to replace a part of the Kalman gain calculations with a neural network. The work can be useful, although does not seem to be too difficult in the presence of other work on matrix manipulations with neural networks. ********************************************* Paper: 347463 Title: Small Digit Capacity Arithmetic for Problems of Discrete Optimization -------------------- review 1 -------------------- ---------------------------- REVIEW 1 -------------------------- PAPER: 7 TITLE: Small Digit Capacity Arithmetic for Problems of Discrete Optimization OVERALL RATING: 1 (weak accept) REVIEWER'S CONFIDENCE: 4 (expert) Relevance to this conference: 4 (good) Originality/Uniqueness: 3 (fair) English readability: 4 (good) Paper organization/presentation: 4 (good) Has good survey been done?: 2 (poor) The title is not appropriate since it does not clarify teh real content of teh paper. Authors should describe better the influence of clipping. The problem addressed by teh paper should be clearly stated in the introduction. Comparisons and discussion of the results should be extended. -------------------- review 2 -------------------- ---------------------------- REVIEW 2 -------------------------- PAPER: 7 TITLE: Small Digit Capacity Arithmetic for Problems of Discrete Optimization OVERALL RATING: 2 (accept) REVIEWER'S CONFIDENCE: 2 (medium) Relevance to this conference: 4 (good) Originality/Uniqueness: 4 (good) English readability: 4 (good) Paper organization/presentation: 3 (fair) Has good survey been done?: 4 (good) Approach could be generally applicable to understanding of effectively of modification clipping procedure with q parameter ********************************************* Paper: 347582 Title: Bayesian regularization of function approximation using orthogonolized basis -------------------- review 1 -------------------- ---------------------------- REVIEW 1 -------------------------- PAPER: 8 TITLE: Bayesian regularization of function approximation using orthogonolized basis OVERALL RATING: 2 (accept) REVIEWER'S CONFIDENCE: 4 (expert) Relevance to this conference: 4 (good) Originality/Uniqueness: 4 (good) English readability: 4 (good) Paper organization/presentation: 4 (good) Has good survey been done?: 4 (good) Comparisons with the literature should be presented more in detail. Results should be discussed more. -------------------- review 2 -------------------- ---------------------------- REVIEW 2 -------------------------- PAPER: 8 TITLE: Bayesian regularization of function approximation using orthogonolized basis OVERALL RATING: 2 (accept) REVIEWER'S CONFIDENCE: 3 (high) Relevance to this conference: 4 (good) Originality/Uniqueness: 3 (fair) English readability: 3 (fair) Paper organization/presentation: 3 (fair) Has good survey been done?: 4 (good) The heart of the problem of scalar function approximation is to build a functional dependence which is able to produce the given dataset. Functional approximation, as most of the inverse problems, is ill-posed. In this paper, a fair survey of the problem is given, first. A point-by-point approximation method for multidimensional scalar function is discussed. And the orthogonalized basis is used for Bayesian regularization of function approximation. The proposed approximation algorithm provides unique analytical solution for the regularization parameters. This study is relevant to this conference. However, the English writing in the paper needs to be further improved or correct. ********************************************* Paper: 347903 Title: The Instrumental System for Face Recognition Tasks -------------------- review 1 -------------------- ---------------------------- REVIEW 1 -------------------------- PAPER: 9 TITLE: The Instrumental System for Face Recognition Tasks OVERALL RATING: 2 (accept) REVIEWER'S CONFIDENCE: 3 (high) Relevance to this conference: 4 (good) Originality/Uniqueness: 4 (good) English readability: 4 (good) Paper organization/presentation: 4 (good) Has good survey been done?: 4 (good) In this paper an upright front face recognition system is presented, aiming to recognize faces on machine readable travel documents (MRTD). In order to increase detection speed and hit rate, the authors introduce some heuristics. Furthermore, for recognition purposes eigenface approach has been introduced. My overall impression is that the proposed work is very well presented. The proposed techniques are novel and sophisticated. The result is a very good paper, that in my personal opinion would be of great interest to research thatdeal with face recognition systems. Concluding, I suggest to be accepted as it is. ********************************************* Paper: 349249 Title: KEYBOARD FINGER GESTURE RECOGNITION USING TIME DELAY NEURAL NETWORKS -------------------- review 1 -------------------- ---------------------------- REVIEW 1 -------------------------- PAPER: 10 TITLE: KEYBOARD FINGER GESTURE RECOGNITION USING TIME DELAY NEURAL NETWORKS OVERALL RATING: 2 (accept) REVIEWER'S CONFIDENCE: 2 (medium) Relevance to this conference: 4 (good) Originality/Uniqueness: 4 (good) English readability: 3 (fair) Paper organization/presentation: 3 (fair) Has good survey been done?: 4 (good) The Author(s) provide(s) both interesting and original approach to gesture recognition. Even though there are some issues that I would like to address to him/her: has he/she considered a situation of using gloves, extensive jewelry or manicure – these would disturb the first step of recognition (skin color model). I also think I would be valuable to compare the problems in hand gestures recognition and face retrival, certainly it could lead to interesting conclusions. Unfortunately Author did not provide enough numerical results that could help to estimate the rate of his/her scientific success (of course I am aware that the survey is still in progress). The paper is prepared correctly with proper information structure, although it lacks images of examples of gestures that could be recognized. Some minor grammar and syntax errors are present but generally I consider the paper valuable and recommend its acceptance. ********************************************* Paper: 349269 Title: Toward a Moving Object Identification by Spiking Neurons -------------------- review 1 -------------------- ---------------------------- REVIEW 1 -------------------------- PAPER: 11 TITLE: Toward a Moving Object Identification by Spiking Neurons OVERALL RATING: -1 (weak reject) REVIEWER'S CONFIDENCE: 2 (medium) Relevance to this conference: 4 (good) Originality/Uniqueness: 2 (poor) English readability: 4 (good) Paper organization/presentation: 3 (fair) Has good survey been done?: 1 (very poor) In general, the paper is written poorly, that was stated by its authors. The question arises, why we should use neural network (in particular spiking neural networks) to solve such problem (there are a lot of methods to identify moving object). There are two obvious answers, one is because they give a significant advantage compared with other methods (not shown in the paper), and the second is that it helps to facilitate interaction with other neural networks (also not reflected in the paper). There is no theoretical or experimental confirmation of the effectiveness of the proposed method for moving object identification. The neural network (section 2.3) used by authors is described quite blurry, resulting in doubts whether it can identify a moving object at all. I'd recommend to accept the report only if authors give at least theoretical or experimental (better both) justification of the effectiveness of proposed method. Article written at fairly good English, and in this respect, there's no any reproaches. -------------------- review 2 -------------------- ---------------------------- REVIEW 2 -------------------------- PAPER: 11 TITLE: Toward a Moving Object Identification by Spiking Neurons OVERALL RATING: 2 (accept) REVIEWER'S CONFIDENCE: 3 (high) Relevance to this conference: 5 (excellent) Originality/Uniqueness: 5 (excellent) English readability: 4 (good) Paper organization/presentation: 4 (good) Has good survey been done?: 3 (fair) The submitted article is a position paper and the author explains capabilities of spiking neurons to be applied in a moving object identification system. Generally speaking, the paper starts on an current example of such system and moves to a novel proposition. It is hard to address the scientific quality as it is a position paper, however due to an interesting application and author's determination to explain the future research project and to discuss it during the conference I accept this paper. Examining the paper's structure - it is well written but the introduction section misses a part explaining the idea of spiking neurons. They occur very rarely in the other papers thus an draft introduction to this unobvious material is a must, otherwise a reader is forced to exploit this subject elsewhere. I recommend to remove the sentence "The author know that even as a position paper, this is totally immature to submit to an international conference." (Final remarks). I'd be more appropriate to introduce a solid future plan instead of excusing oneself, especially that the article, as quite interesting proposition, should be credited to be publicly discussed. From the editorial perspective, there are no major errors apart that the equations numbers are missing. -------------------- review 3 -------------------- ---------------------------- REVIEW 3 -------------------------- PAPER: 11 TITLE: Toward a Moving Object Identification by Spiking Neurons OVERALL RATING: -3 (strong reject) REVIEWER'S CONFIDENCE: 4 (expert) Relevance to this conference: 1 (very poor) Originality/Uniqueness: 1 (very poor) English readability: 4 (good) Paper organization/presentation: 1 (very poor) Has good survey been done?: 1 (very poor) The paper presents - as the authors state - only a position paper. There is no research presentation, no result, nothing. Basically is the presentation of a known problem and some background knowledge which is already available in the literature. There is no value even from the point of view of a tutorial or as introduction to the problem. ********************************************* Paper: 349323 Title: The Hybrid Feature Selection for the Prediction of Household Bankruptcy -------------------- review 1 -------------------- ---------------------------- REVIEW 1 -------------------------- PAPER: 12 TITLE: The Hybrid Feature Selection for the Prediction of Household Bankruptcy OVERALL RATING: 1 (weak accept) REVIEWER'S CONFIDENCE: 2 (medium) Relevance to this conference: 3 (fair) Originality/Uniqueness: 2 (poor) English readability: 2 (poor) Paper organization/presentation: 3 (fair) Has good survey been done?: 3 (fair) This paper proposes a hybrid algorithm called Multi Measurement and Voting for reducing the risk of selecting non-optimality features. Several popular algorithms of the feature selection, i.e. InfoGain, GainRatio and Chi2, are combined to construct the hybrid algorithm. The work reported in the paper has nothing new in the algorithms for feature selection. Furthermore, the experiment results of the algorithm lack evaluation. And the English writing of the paper is poor. ********************************************* Paper: 349393 Title: Topological Organization for Categorical Data Clustering -------------------- review 1 -------------------- ---------------------------- REVIEW 1 -------------------------- PAPER: 13 TITLE: Topological Organization for Categorical Data Clustering OVERALL RATING: 3 (strong accept) REVIEWER'S CONFIDENCE: 2 (medium) Relevance to this conference: 5 (excellent) Originality/Uniqueness: 4 (good) English readability: 5 (excellent) Paper organization/presentation: 5 (excellent) Has good survey been done?: 5 (excellent) In the paper the Authors address very interesting and important problem from pattern recognition domain – categorical data clasterization and visualization. The Authors are undoubtedly very well prepared to process such a research – they prove it in the first section that contains state of art and current methods comparison. The Relational Analysis method they had developed and introduced in the paper is valuable and significant, yet more interesting because of using heuristic approach. The whole presentation is constructed understandably with great attention to details and the performance results are attached as well. It was very honest to remark possible drawbacks of the new approach. The paper is prepared correctly with proper information structure, English readability is very good. -------------------- review 2 -------------------- ---------------------------- REVIEW 2 -------------------------- PAPER: 13 TITLE: Topological Organization for Categorical Data Clustering OVERALL RATING: 2 (accept) REVIEWER'S CONFIDENCE: 2 (medium) Relevance to this conference: 4 (good) Originality/Uniqueness: 4 (good) English readability: 4 (good) Paper organization/presentation: 3 (fair) Has good survey been done?: 3 (fair) I am just wondering if the problem can be solved equally by using k-means; and if yes, whether the proposed method is really better, in what sense. ********************************************* Paper: 350639 Title: Parallel and Distributed Evolutionary Computation: A Survey -------------------- review 1 -------------------- ---------------------------- REVIEW 1 -------------------------- PAPER: 14 TITLE: Parallel and Distributed Evolutionary Computation: A Survey OVERALL RATING: 0 (borderline paper) REVIEWER'S CONFIDENCE: 2 (medium) Relevance to this conference: 4 (good) Originality/Uniqueness: 2 (poor) English readability: 3 (fair) Paper organization/presentation: 2 (poor) Has good survey been done?: 3 (fair) COMMENTS: 1. The paper provides a fair review of the parallel and distributed evolutionary computation. 2. The comparison part of the presented approaches are missing. 3. The order of the figures is irregular, i.e. the order begins from Fig. 4.1 of section 2.2. 4. The paper only refers to the citations no. 4, 5, 6, and 22 from the references; the rest are not been cited in the paper. -------------------- review 2 -------------------- ---------------------------- REVIEW 2 -------------------------- PAPER: 14 TITLE: Parallel and Distributed Evolutionary Computation: A Survey OVERALL RATING: 0 (borderline paper) REVIEWER'S CONFIDENCE: 3 (high) Relevance to this conference: 4 (good) Originality/Uniqueness: 2 (poor) English readability: 4 (good) Paper organization/presentation: 3 (fair) Has good survey been done?: 4 (good) The paper is a survey. It does not include original results. The format of the paper is not correct. There are many references that are not mentioned in the paper. Numeration of figures is not correct. Only the topic of the paper is interesting, so the paper can be accepted only after radical improvement. Final conclusion: the paper can be accepted only after radical improvement. ********************************************* Paper: 351456 Title: A Multi-Experts Methodology for Early Detection and Diagnosis of Mechanical Defects -------------------- review 1 -------------------- ---------------------------- REVIEW 1 -------------------------- PAPER: 15 TITLE: A Multi-Experts Methodology for Early Detection and Diagnosis of Mechanical Defects OVERALL RATING: 3 (strong accept) REVIEWER'S CONFIDENCE: 3 (high) Relevance to this conference: 4 (good) Originality/Uniqueness: 3 (fair) English readability: 5 (excellent) Paper organization/presentation: 5 (excellent) Has good survey been done?: 4 (good) Very well-written paper. My only remark is about figures - fig. 2, 3 and 8 are not readable and should be enlarged. -------------------- review 2 -------------------- ---------------------------- REVIEW 2 -------------------------- PAPER: 15 TITLE: A Multi-Experts Methodology for Early Detection and Diagnosis of Mechanical Defects OVERALL RATING: 2 (accept) REVIEWER'S CONFIDENCE: 2 (medium) Relevance to this conference: 4 (good) Originality/Uniqueness: 4 (good) English readability: 3 (fair) Paper organization/presentation: 4 (good) Has good survey been done?: 4 (good) The paper presents original neural based multi-expert architecture for early detection of mechanical defects. It also provides links to the similar works. I am not an expert in English, but in my opinion article reads pretty hard, I believe that many expressions could be simplified. -------------------- review 3 -------------------- ---------------------------- REVIEW 3 -------------------------- PAPER: 15 TITLE: A Multi-Experts Methodology for Early Detection and Diagnosis of Mechanical Defects OVERALL RATING: 2 (accept) REVIEWER'S CONFIDENCE: 3 (high) Relevance to this conference: 4 (good) Originality/Uniqueness: 3 (fair) English readability: 4 (good) Paper organization/presentation: 4 (good) Has good survey been done?: 4 (good) The paper presents a multi-expert method for mechanical defect detection. It uses some practical/real data, which is good. But it is not clear how novel the method is. The paper can be accepted if the following minor changes are made: it would be beneficial to mention more relevant work in detection with soft computing techniques, for example, V. Mitra, Chia-Jiu Wang, and S. Banerjee, "Lidar detection of underwater objects using a neuro-SVM-based architecture," IEEE Transactions on Neural Networks, no.17, pp.717-731, 2006. L.P. Wang and X.J. Fu, Data Mining with Computational Intelligence, Springer, Berlin, 2005. ********************************************* Paper: 351461 Title: ANFIS controller for nonholonomic robots -------------------- review 1 -------------------- ---------------------------- REVIEW 1 -------------------------- PAPER: 16 TITLE: ANFIS controller for nonholonomic robots OVERALL RATING: 2 (accept) REVIEWER'S CONFIDENCE: 3 (high) Relevance to this conference: 4 (good) Originality/Uniqueness: 4 (good) English readability: 4 (good) Paper organization/presentation: 4 (good) Has good survey been done?: 4 (good) The paper is good. The topic of the paper is interesting. However, the method of research and results of simulations can be explained more clearly. Some technical recommendations are below. 1. References should be made in accordance with format requirements of the ICNNAI-2010 (more correctly). 2. There are several small mistakes in equations and in the paper text. These mistakes should be removed. In particular, English of the text should be checked. 3. In accordance with format requirements of the ICNNAI-2010, equations should be mentioned in the text as Eq. (1) or Eqs. (2),(3). Final conclusion: the paper can be accepted after small improvement. -------------------- review 2 -------------------- ---------------------------- REVIEW 2 -------------------------- PAPER: 16 TITLE: ANFIS controller for nonholonomic robots OVERALL RATING: -1 (weak reject) REVIEWER'S CONFIDENCE: 2 (medium) Relevance to this conference: 4 (good) Originality/Uniqueness: 2 (poor) English readability: 3 (fair) Paper organization/presentation: 3 (fair) Has good survey been done?: 3 (fair) There are many papers discussing robot control using neuro-fuzzy networks. In particular, the paper did not reference the following paper: Autonomous parallel parking of a car-like mobile robot by a neuro-fuzzy behavior-based controller Khoshnejad, M.; Demirli, K.; Fuzzy Information Processing Society, 2005. NAFIPS 2005. Annual Meeting of the North American Publication Year: 2005 , Page(s): 814 - 819 ********************************************* Paper: 353388 Title: Ophthalmologic Image Segmentation and Surface Visualization -------------------- review 1 -------------------- ---------------------------- REVIEW 1 -------------------------- PAPER: 17 TITLE: Ophthalmologic Image Segmentation and Surface Visualization OVERALL RATING: 2 (accept) REVIEWER'S CONFIDENCE: 3 (high) Relevance to this conference: 4 (good) Originality/Uniqueness: 3 (fair) English readability: 4 (good) Paper organization/presentation: 4 (good) Has good survey been done?: 1 (very poor) The subject is interesting and the paper also but it is not finished. It covers only the description of the method but there are non experimental results. The paper should be published but after completing section "Results". ********************************************* Paper: 354255 Title: A New Validity Measure for Heuristic Possibilistic Clustering -------------------- review 1 -------------------- ---------------------------- REVIEW 1 -------------------------- PAPER: 18 TITLE: A New Validity Measure for Heuristic Possibilistic Clustering OVERALL RATING: 0 (borderline paper) REVIEWER'S CONFIDENCE: 3 (high) Relevance to this conference: 4 (good) Originality/Uniqueness: 3 (fair) English readability: 3 (fair) Paper organization/presentation: 3 (fair) Has good survey been done?: 3 (fair) The new validity measure seems interesting. However, just one dataset cannot draw any conclusion. You should use at least 3-5 datasets to verify the proposed method. There are some minor grammatical errors. Please polish the English if the paper is accepted. -------------------- review 2 -------------------- ---------------------------- REVIEW 2 -------------------------- PAPER: 18 TITLE: A New Validity Measure for Heuristic Possibilistic Clustering OVERALL RATING: 1 (weak accept) REVIEWER'S CONFIDENCE: 2 (medium) Relevance to this conference: 3 (fair) Originality/Uniqueness: 3 (fair) English readability: 3 (fair) Paper organization/presentation: 3 (fair) Has good survey been done?: 3 (fair) In this paper, a validity measure for the heuristic D-AFC(c) algorithm of possibilistic clustering is proposed. However, in the numerical experiment only one example is given, which is not enough to clarify the problem. In addition, the English writing needs to be more correct and improved.