Science topic

Network Science - Science topic

Physical, engineered, information, biological, cognitive, semantic and social network research.
Questions related to Network Science
  • asked a question related to Network Science
Question
7 answers
what are  the key differences between CCN and NDN in information centeric networks?
Relevant answer
Answer
NDN architecture is an extended project, which is funded by National Science Foundation (NSF) program. It is an improved version of the CCN. Similar to CCN, NDN also follows the Interest packet and Data packet order to receive any appropriate content. However, there are some structural differences built into the NDN that reduces the Interest packet and Data packet search time and the issue of Interest packet looping. This architecture is a report as drafted by Zhang et al in 2010 as the blueprint of the general NDN concepts, its architecture and stated that “NDN is a global overlay” as IP. This is because the hierarchically developed structured names in NDN easily correspond to the hierarchically developed Structured Address of IP and "BGP, OSPF, IS-IS", which are the basic routing protocols IP that can be easily used to propagate NDN over and in parallel with IP
  • asked a question related to Network Science
Question
9 answers
Is there a Python or R package for analyzing spreader nodes and community detection in the multilayer network?
Relevant answer
Answer
I worked on this specific topic in my dissertation, I tried many tools but eventually, I and my research team decided to design a simulator for static, dynamic, and multilayer networks using NetLogo multiagent programming. Below is the link to the simulator, it might be useful in your work.
Note that you just need to configure the simulator based on your needs.
Good Luck
  • asked a question related to Network Science
Question
4 answers
Thresholding is a common step in the preparation of a connectivity matrix for graph-theoretical analysis. This is done by applying an absolute or a proportional weight threshold, but this threshold is often arbitrarily determined. Most papers I found do use a threshold, but I saw at least a few that didn't.
  • Given a functional connectivity matrix based on EEG data (the measure used to compute connectivity is the phase-lag index (PLI), an index of phase synchronization), is thresholding an essential step? Or can graph-theoretical measures be computed also on the completely connected, weighted graph?
Some papers used a range of proportional thresholds (e.g. 10% -90% of the strongest connections, in steps of 2.5%), obtaining a graph for each threshold, then computed a graph-theoretical measure for each graph and then averaged these measures across the different graphs to get a single measure. Others, instead of a threshold, used a minimally spanning tree (MST) to retain the strongest connections.
  • If thresholding is the best practice, what is the best approach to do so? What are the pros and cons of each?
Relevant answer
Answer
Hi, Marta
It's very interesting topic about the threshold choice during the brain network analysis. I was also concerned this threshold problem before. But now I have read a lot of papers and got some inspirations, here I share with you!
(1) First, as Avraam answered, minimum spanning tree (MST) is an unbiased choice to avoid the threshold as the MST extract the backbone of the network and it generates the non-cycle graph with all nodes connected (also with total smallest edge distance). CJ Stam and his team have proved the MST is another alternative choice to avoid the threshold. Maybe you can search the google scholoar by using the key words "CJ Stam", "minimum spanning tree", "unbiased" or other related key words. I have also focused on the minimum spanning tree in EEG brain network and recent I am ready to submit a paper about this topic, if someday the paper was accepted and published, I can send it to your email. Rather than traditional MST, I have also given some new try in this paper.
(2) Second, maybe you can try the fully connected weighted network (rather than binary), that is no threshold used. Although fully connected network may not represent the true brain network connectivity state, however, it provides very robust classification features to distinguish the EEG resting and task states for both healthy subjects and patients. you can read my recent published paper " Changes in Brain Functional Network Connectivity in Adult Moyamoya Disease". In this paper, we adopted the fully connected weighed network rather than chose a fixed threshold, maybe you can try this way.
(3) Third, the another automatic threshold choice strategy is based on the statistical significance. I read this paper "A novel index of functional connectivity: phase lag index based on Wikcoxon signed rank test". Although it's a method related paper, it inspired me to consider using the statistical test to automatically avoid the threshold. You can download it and hope it be helpful to you.
If you still have any question about the brain network analysis, you can send your puzzle by email "gxzheng16@fudan.edu.cn". If I have spare time, and I will give your reply.
Gaoxing Zheng
  • asked a question related to Network Science
Question
5 answers
Dear all,
I would like to know if it is possible to use SAOMs (Stochastic actor oriented models) to analyse weighted networks?
Thank you in advance,
Léa DAUPAGNE
Relevant answer
Answer
There seems to be a less ised feature to handle ordinal values. It is discussed here: http://lists.r-forge.r-project.org/pipermail/rsiena-help/2013-July/000272.html
  • asked a question related to Network Science
Question
4 answers
Dear all,
I would like to know if it is currenly possible to use temporal ERGMs (Exponential Random Graph Models) for analyzing weighted networks?
For now, it seems that software packages available to analyse TERGMs (tergm or btergm) only use binary networks.
Thanks in advance for your answer,
Léa Daupagne
  • asked a question related to Network Science
Question
2 answers
I want to consider the effect of layers similarity to solve a problem in multilayer networks. Can anyone help me which are the most important interlayer similarities in multilayer networks?
Relevant answer
Answer
My goal is to study the similarity famous between layers in the multiplex networks like overlap links. But I do not know which exactly the similarities between the layers are very important.
  • asked a question related to Network Science
Question
8 answers
What are the ways to transfer a graph from one Relation space to a Euclidean space with less time complexity? although there are some ways solution (such as signal process, spectral method ), they have a high time complexity.
Relevant answer
Answer
Dear Kamal,
maybe node2vec can be useful for your application: https://snap.stanford.edu/node2vec/
Kind regards,
Djordje
  • asked a question related to Network Science
Question
8 answers
I want to convert an unweighted graph to weighted for solving the link prediction problem. Is the best way to transfer from an unweighted graph to a weighted graph to consider the similarity between nodes?
Relevant answer
Answer
Not necessarily. It totally depends on your application. If your dataset is on online social networks and you want to model the relation strength among individuals, you could also consider the degree of intimacy, trustworthiness, and influence among individuals. check out these papers:
For the concept of weight on multiplex networks:
Also take a look at Granovetter paper since it is possibly the first paper who defined the concept of weak and strong ties in social networks and modeled them as a set of nodes and links: https://sociology.stanford.edu/sites/g/files/sbiybj9501/f/publications/the_strength_of_weak_ties_and_exch_w-gans.pdf
  • asked a question related to Network Science
Question
1 answer
I am interested in creating a multi-layer mechanical network. Therefore I would like to find a software where you can visualise nodes and links moving around in 2D and 3D space.
Relevant answer
Answer
Thank you
  • asked a question related to Network Science
Question
6 answers
As we know that communities size have power-law distribution. Is there a relationship between power-law distribution of communities size and communities hierarchical structure? Can it be said that communiteis locating the last part of the power-law distribution have a stronger hierarchical structure?
Relevant answer
Answer
Hierarchy and size are two different independent aspects.
Also hierarchy can not be labelled as strong or weak......
  • asked a question related to Network Science
Question
4 answers
We are organizing, together with Matthieu Gilson, Adrià Tauste and Gorka Zamora-López, a Hands-on course on neural data science, in the frame of the XIII Summer School in Statistics at UPC-UB (Barcelona, Spain).
The course will cover statistics, time series modelling, machine learning and graph theory. The students will develop a data-driven project throughout the course to understand how these different tools can work together to analyse brain data. Reach out if you have any question!
When: July 1st to 5th from 3:00 PM to 6:00 PM
Where: Barcelona, UPC campus
Relevant answer
Answer
Thanks for sharing
  • asked a question related to Network Science
Question
2 answers
There is a question involved me a few days ago. Is there a correlation between power-law distribution and the Pareto principle ( 80/20 rule ) in natural phenomena?Are they describe the one thing?
Relevant answer
Answer
same
  • asked a question related to Network Science
Question
5 answers
Complex networks have the number of common features, but is there an outstanding feature of each complex network which does not exist in the rest? For instance, the Rich-club property, there is more to be seen in the brain networks.Can we say each of them has a unique feature?
Relevant answer
Answer
Depending on the class of these complex networks. Scale-free networks for example share a common property of having a degree distribution that follows power-law distribution.
Regarding the rich-club coefficient: From my research on social crowdsourcing communities (SCCs), I found that their networks are scale-free and have small Rich-club coefficients, contrary to social networks who are scale-free and have higher rich-club coefficients.
The small rich-club coefficient makes makes sense in SSc because they have moderators who tend to very active and highly popular; yet rarely associate with each other (through following or befriending) or interact with each other in the SCC. This is because of the nature of their duties that makes it useless for them to associate with one another. In my case specifically, moderators are hired by the same SCC so they work off of the same physical space and therefore have direct access to each other off-line, which limits the need for on-line association.
  • asked a question related to Network Science
Question
23 answers
Please i will use Network DEA to Education sector, do you know the software that i can use for NeworK DEA model.
Thank you so much
Relevant answer
Answer
I saw that this web site dont have the full text. What your email that i can send to you?
  • asked a question related to Network Science
Question
9 answers
I know NS and GloMoSim for Ad-hoc networks. Is there a simulator used for home networks?
Relevant answer
Answer
Defiantly, Cisco Packet Tracer is very good solution for Home Networks
You can checkout the following tutorials:
  • asked a question related to Network Science
Question
2 answers
field: traffic, flood, structure, community resilience, hazard,sociology, network science
pls contact kfeng46@gatech.edu
Relevant answer
Answer
I have a paper about flood impacts on road transportation. Let me know if you are interested in having a copy:
  • asked a question related to Network Science
Question
5 answers
I am doing M.Tech in Mathematics and Computing from IIT Patna and I want to go for Ph.D. after M.tech. I am from Mathematics background (M.Sc.). I want to know the research opportunities in the fields of network science and artificial intelligence.
Relevant answer
Answer
Hey there,
You can find interesting topics, related to artificial intelligence, in "https://www.quora.com/What-are-the-hot-topics-in-artificial-intelligence-for-research". In general, neural networks (in particular, convolutional neural networks) have received many efforts of the research community, therefore, this is an important/relevant topic.
  • asked a question related to Network Science
Question
5 answers
I'm going to construct  knowledge models of concepts in complex knowledge domains.
It seems that a multi-layer network representation of knowledge could offer a richness that is lacking in a monoplex representation. Concepts are to be expressed as nodes and relations between concepts as edges. In a multi-layer network there are multiple types of relations and thus edges between nodes. In case you are not clear on what I mean by multi-layer network, here are links to papers called Mathematical Formulation of Multilayer Networks and An Introduction to Multilayer Networks: https://people.maths.ox.ac.uk/porterm/papers/PhysRevX.3.041022.pdf http://my.fit.edu/~mtomasini/Resources/multilayer-networks.pdf.
Metrics for the analysis of monoplex networks can be transferred to multi-layer networks. Thus, with my model I could calculate the community structure of some cluster of concepts or calculate the betweeness centrality of a given concept. The model may have relevance for personal knowledge management \ education, scientific analysis of knowledge, and as a knowledge graph powering a knowledge engine. I'm motivated to, of course, map my own knowledge as I learn. Also, I would like to make a highly interactive program that allows an author to make optimal use of available knowledge when composing research.
I plan on using muxViz to construct the networks, analyze, and visualize the data.
Relevant answer
Answer
Hi, maybe multiplex lexical networks might be of relevance here.
Together with my co-authors, we built a multiplex network representation of word-word similarities including different layers/aspects of language such as synonyms/feature sharing/free associations/taxonomical relationships and also other phonological aspects.
In children we found that the multiplex structure has a higher prediction power compared to single layer networks or to other psycholinguistic variables (e.g. frequency, length) in predicting normative age of acquisition, how most English toddlers learn words over time.
In adults we found that the multiplex layers allow to detect a core of densely connected concepts which: (i) emerges suddenly through an explosive phase transition around age 7-8 yrs and (ii) makes the whole mental lexicon highly resilient to random word failures (e.g. concepts becoming progressively inaccessible).
In clinical populations like people with aphasia, we also found that the multiplex closeness of words is an important predictor of performance in a picture naming task (when shown a picture, can people with aphasia name the concept represented in the picture?). This predictive power is absent on the individual layers composing the multiplex lexical network but it appears when multiple semantic and phonological layers are combined together.
Hope these works can be of relevance for your research!
  • asked a question related to Network Science
Question
23 answers
Please help me
Relevant answer
Answer
The approximate number of clusters is calculated using the following formula:
k=the square root of n divided by 2
where k is the number of clusters and n is the cluster size
  • asked a question related to Network Science
Question
5 answers
I am currently analyzing collaboration networks in science. I have used SBM and ERGM to model a cumulative snapshot of the network. I know these methods alone are not sufficient to accurately model my network since it is a longitudinal network. I heard of Stochastic Actor-Based (or actor-oriented) Block Modeling (SABM) which can model my network taking into account it's dynamics in time. I have read some papers on it but have been unable to find a tutorial to guide me through its successful application on my data. Can anyone help? 
Relevant answer
Answer
You might find the lecture and notes from the Social Networks And Health at the Duke Network Analysis Center (DNAC) useful; includes R code, David Schafer did the presentation.
  • asked a question related to Network Science
Question
11 answers
Nowadays complex network data sets are reaching enormous sizes, so that their analysis challenges algorithms, software and hardware. One possible approach to this is to reduce the amount of data while preserving important information. I'm specifically interested in methods that, given a complex network as input, filter out a significant fraction of the edges while preserving structural properties such as degree distribution, components, clustering coefficients, community structure and centrality. Another term used in this context is "backbone", a subset of important edges that represent the network structure.
There are methods to sparsify/filter/sample edges and preserve a specific property. But are there any methods that aim to preserve a large set of diverse properties?
Relevant answer
Answer
I think it really depends if the properties are local or global. In anyway, it seems to be controllable at multiple scales of coarseness. We introduced a multilevel sparsification by algebraic distance, so it partially addresses this question. Please take a look at this paper "Single and multi-level network sparsification by algebraic distance".
  • asked a question related to Network Science
Question
3 answers
Hopefully I have the Forret&Dougherty networking behaviors scale and H-G Wolff questionnaire of networkigng scale? Do you have anything more?
Thanks in advance,
Marzena
Relevant answer
Answer
Thanks. ICT is one of dimension of networking competencies of employees I figured out in research project.
  • asked a question related to Network Science
Question
3 answers
I am looking for research on the common social dynamic known as the "telephone game" — a message is told and re-told, becoming more and more distorted with each re-telling. Specifically, is there telephone game research from a social network analysis perspective?
* Does network density and clustering coefficient matter?
* How do “bridges” and “hubs” affect the message flow?
* How distorted does a message get with each re-telling?
- interested in verbal/written interaction between people, not the forwarding of a static document or email along a chain
Any pointers, ideas?
  • asked a question related to Network Science
Question
3 answers
Bandwidth estimation of packet length based covert chnnel
Relevant answer
Answer
  • asked a question related to Network Science
Question
2 answers
if i click serial monitor in waspmote IDE, i can able to view the output. how to store the files automatically in the text file every hour?
Relevant answer
Answer
When GAMS runs an output file is automatically generated.  That output file has the file extension LST.  The first part of the LST file name is the first part of the name of the GAMS input file.  Thus, if the file myfile.gms is the file run through GAMS the output will be on myfile.LST.  This output file is an ASCII text file in the courier font that can be loaded into a text editor. GAMS also generates a LXI file (in this case myfile.LXI that contains navigation information that is used by the IDE in moving generating the navigation window that permits around the LST file.
Output also appears on the LOG file particularly if one is using the IDE.  That output summarizes the run results and also can serve as a navigation aid when using the IDE.  The LOG file does contain information relevant to the solution and error status.  It is named using the same practice as with the LST file but the file extension is LOG i.e. myfile.LOG.  It also is reproduced in the process window of the IDE.
  • asked a question related to Network Science
Question
3 answers
Hello everyone, 
I am trying to analyze the timing behaviors when a signal is sent between two raspberry pi devices, and trying to use Net Analyzer (Hilscher) for that. But, I am facing few difficulties. If you can help me with this, kindly write back and I will discuss with you in details!
Aranya
Relevant answer
Answer
Hi Mbida,
Thank you for your answer. I have gone through their documentations. But, I have a definite doubt that I think, a person who has got experience with the product, will be able to help me with. 
Aranya
  • asked a question related to Network Science
Question
3 answers
I need to measure the UDP packet losses in a telerobotic Operation. Due to its nature (conectionless) it is not easy to do so. I have tried to do it using Wireshark but is not possible I believe.  Can  anybody help  me please   
 I am trying with Wireshark and what I have realized up to know is that is possible to measure TCP packet losses , but as far as UDP packet losses I don't know how and whether is possible to be done
Relevant answer
Answer
Sir 
You may try using wireshark tool for mesuring the UDP packet loss
  • asked a question related to Network Science
Question
4 answers
 A line of research (e.g. [1]) has considered the question of self-similarity and self-dissimilarity in complex networks.
Very generally, a self-similar object is similar to a part of itself. A natural application of this definition to networks would be to say that a network is self-similar if its graph is composed of subgraphs that are structurally similar to the graph as a whole. Still, this is not yet a precise, quantifiable definition - for instance, what does it mean to say that the graph is "composed" of subgraphs?
Please comment and discuss.
Relevant answer
Answer
Graph composed of subgraphs...
Easy to compose a single graph component from "triangles" -- both open and closed.
When you allow for "overlap" of open and closed triangles, almost an unlimited number of graphs can be created.  Whether they are self-similar is another, more difficult question. Core-periphery structure is often seen in human networks. The core-periphery structures often repeats at different levels of focus/aggregation.
Attached is a graph of an actual business organization showing only Simmelian Ties (Krackhardt, 1998) -- closed triangles of strong ties only.  Notice core-periphery in each cluster and in the overall organization.
  • asked a question related to Network Science
Question
2 answers
I am doing research on user experience with smart AV networks. If anyone has papers, case studies or documents about smart AV Networks, please do share.
Relevant answer
Answer
Smart Audio Visual System/Network...A network system that comprises of smart controlled equipment such as PLCs', display, lighting controls etc. 
  • asked a question related to Network Science
Question
4 answers
What is Network Forensics ? Can anyone recommend some articles or materials.
Relevant answer
Answer
Hello Professor,Network forensics is the process of capturing information that moves over a network
and trying to make sense of it in some kind of forensics capacity. This method is
based on reconstructive traffic analysis. It could be used for forensic analysis to read
and analyze the contents of the Internet raw data in PCAP format for a particular
session on the network.
I have bben in network forensic research field for the past 10 years. feel free to ask any doubts
  • asked a question related to Network Science
Question
2 answers
I'm currently working on a research on User-centric wireless network and need more insight into to the architecture in order to be able to simulate the network for my research.
Relevant answer
Answer
It is an area of research where the contents might be recommended to the users based on their profiles and their surfing activities. Different architectures can be proposed based on the issues to be addressed regarding User-Centric Networks and the approach to be taken to address the particular issues. I suggest the following reading list.
1. Xiaoshuang Xing ,Tao Jing ; Wei Zhou ; Xiuzhen Cheng ; Yan Huo ; Hang Liu, Routing in user-centric networks, IEEE Communication Magazine, Vol 52, Issue 9, pp. 44-51, September 2014.
  • asked a question related to Network Science
Question
9 answers
I'm trying to choose my MSc research area in the field of social network analysis but I'm new to this subject and am not aware of currently active research areas or the trends. I would appreciate any suggestion you might give, so I can choose the one closest to my interests.
Relevant answer
Answer
Go for sentiment analysis ...specially for none English language ..this field is so interesting for social media analysis 
Rgds 
  • asked a question related to Network Science
Question
4 answers
Once an information flow (IF) can be represented as a process, can we measure your complexity as a network?
According Brandes, Robins, McCranie & Wasserman (2013), in order to think in terms of network first is necessary to have elements (E) and processes (P) of a network model: phenomenon (E), which passes through an abstraction (P), making the concept of the (E) which may be represented (P) and data network (E).
Taking the IF as a broad phenomenon, can we introduce the concept of network processes, which can be viewed as a web of interdependent tasks or activities, its products, which are more than the sum of its components and the participating organizations, can be understood as a network of people and communications?
Which measure can be more appropriate in this case?
Relevant answer
Answer
As in an complexity measure, you must first consider the usefulness of the metric. In essence, a metric should preserve relations in the real world (objective or intersubjective) in such a way that if X has relation R to Y in the real world it is reflected in the metric. It should also be possible to make informed decisions based on the metric in an activity performed by an actor that tries to achieve a goal. 
So, what kind of activities do you envision such a metric to be useful in? What kind of decisions do you want to make (e.g., alternatives) based on this metric? If you fail to envision this, then what good would a metric do? I can define "squarishness" as a criteria and arbitrarily start counting squarish objects in a region, but to what avail I do not know. The metric can be made both correct and meaningful, but hardly useful (unless I find a reason for counting squarish objects).
  • asked a question related to Network Science
Question
15 answers
This can be seen in many disciplines under different terminology. I need to get get as much ideas as possible. Feel free to let me know how it is handled in realms closer to your research. For instance
  • Sociophysics: Opinion formation on social networks 
  • Marketing: viral marketing 
  • Microfinance : Diffusion models to study the influential individuals, etc...
In brief the idea is: given a network, if we need to drive an idea, which node (or agent) we should select. All your comments are highly appreciated. Thank you in advance.
Relevant answer
Answer
You need to evaluate authority and hub centrality.  
  • asked a question related to Network Science
Question
10 answers
I've seen networks of this type used to highlight communities within your list of friends on Facebook. It removes yourself from the network, because this node is by default connected to everyone and clutters the visualization.  Looking to see if there is a standard technical term for such a network.
Relevant answer
Answer
Ego's alter network works in academia.  With business clients I use Ego's "network neighborhood" -- a more common term and because there are both direct and indirect alters.  Here is a picture of Erdøs' famous network neighborhood... with Paul Erdøs removed.
  • asked a question related to Network Science
Question
337 answers
We have recently seen a lot of down-votings in some threads. We also saw different kinds of reactions on this phenomenon. I would like to ask a question about these reactions.
First I assume that there are several possible reasons for downvotings: dissent – misunderstandings – misuse of buttons without knowing it– a social scientist who writes a paper about the reactions – a test carried out by RG – some technical problem, etc.
I would like to have a discussion about the reactions on the part of researchers. We have seen calls to ban down-voters or to cancel their anonymity. Elections in democracies are anonymous and there are good reasons for this. I think most of us agree about some basic traits of democracy, the right to stay anonymous is among them. What about these basics in social media?
Relevant answer
Answer
I don't know why so many downvotes without reasonable explanation? and How I can find a researcher that downvote my answer and question?
Please see this question and answer:
  • asked a question related to Network Science
Question
7 answers
I am trying to understand how a complexity (quantified) can be attributed to any virtual path from an attribute to another attribute of a different entity.
Relevant answer
Answer
Formal models are good for estimating the development effort, but the model is completed late in the process (waterfall, scrum). Generally, management wants an effort estimate that included the modeling or design process. In essence, the question is what will it cost to provide the require facilities, This is why the point techniques are favored by many developers.
  • asked a question related to Network Science
Question
7 answers
Hi friends,
After your experiences, tell me what the best position of the base station in an area of 100 x 100 m, is the center, or the extremity or outside corresponding to the best density (number of nodes)?
Thank you.
Best regards
Relevant answer
Answer
All WSN papers that deal with clustering locate sink in the center. The locations of other nodes are random with normal distribution. this makes the center more near to most of the nodes. If you try to use K-mean algrithm to find four different clusters you will find that they located around the center. 
  • asked a question related to Network Science
Question
7 answers
I used nnet package in R to train the neural network and make prediction. At first, because the output values were large, i used the formula (x-xmin)/(xmax-xmin) to standardize them in range of 0 to 1. After training the network, i predicted the output values. The result is a range of data in range of 0 and 1. How can i un standardize the predicted values to have predicted values of the first unit?
thanks 
Relevant answer
Answer
Dear Ghazaleh,
here are some suggestions based on the information you provided.
As the values of input variables are a mixture of binary (hence, they should belong to range [0,1]) and quantitative currency (hence, they should belong to to range like [1000$, 1000000$])  I suggest to scale the input variables so that all ones belong to the same range [0,1]; so it means applying the transformation (X-mean(X))/sd(X) to all input variables
If you use caret you can just specify this transformation inside the the pre-process object. For example, if XX is your train set (not scaled) of 1980 observations, yy the related response variables (not scaled), and ZZ the test set (not scaled) of 540 observations,      
fit <- train(y = yy, x = XX ,
method = "avNNet",  
preProc = c("center", "scale"),
tuneGrid = expand.grid(.decay = c(0.001, .01, .1), .size = seq(1, 27, by = 2), .bag = FALSE), 
linout = TRUE, trace = FALSE, maxit = 1000,
trControl = trainControl(method = "repeatedcv",repeats = 5, number = 10)) 
When you make the prediction  - myPred = predict(fit,ZZ) -  you don't need to scale the test set, as caret scales it for you.  
If nnet still coverges pretty quicly with this code, I would suggest 3 transformations on the train set and test set on cascade, i.e. applying i+1 transformation only if the problem is still not fixed with the i transformation. These transformations are very easy to do with caret. 
1) removing near zero var predictors
data = rbind(XX,ZZ)
PredToDel = nearZeroVar(data)
data = data [,-PredToDel]
XX = data[1:1980,]
ZZ = data[1981:nrow(data),]
2) removing predictors that make ill-conditioned square matrix
data = rbind(XX,ZZ)
PredToDel = trim.matrix( cov( data ) )
XX = data[1:1980,]
ZZ = data[1981:nrow(data),]
3) removing high correlated predictors
data = rbind(XX,ZZ)
PredToDel = findCorrelation(cor( data ))
XX = data[1:1980,]
ZZ = data[1981:nrow(data),]
Hope this can help you  
  • asked a question related to Network Science
Question
7 answers
i would like to know different parameter of vampire attack on nodes.
Relevant answer
Answer
Vampire Attacks parameters: Stretch attacks create a long route(false Route) from Source to Destination which consumes lot of energy. For Example Valid route(Path) from Source to destination A-->C === A-->B-->C. But false route will be some thing like A-->C === A-->B-->D. -->E-->F-->C. Parameters are Source, Destination, Packet Id, Time to live, Next Hop, Routers etc. The parameters mainly depend on the specific Network. The parameters that are mentioned above are common to all.
  • asked a question related to Network Science
Question
3 answers
My work requires the Real-world data sets for arrival of tasks and task service time.
Relevant answer
Answer
Could you please see our paper on Job Scheduling for Cloud Computing Using Neural Networks. It might help. Yours,
  • asked a question related to Network Science
Question
6 answers
Is anyone working on Contact Tracing, in the current Ebola outbreaks, using non-interview (non-F2F) techniques (such as tracking mobile phones) to gather information on movements/interactions of infectious individuals? 
  • asked a question related to Network Science
Question
15 answers
There is basically no limit to the phenomena that can be modeled and analyzed in terms of complex networks - entities and their relationships between which can be represented as the nodes and edges of a graph, and which form a non-trivial pattern. So let's make this a small survey:
  • Where in your research do you employ complex networks and network analysis methods?
  • What are your data sources? How big are they?
  • Which tools do you use for the network analysis process?
  • What did you learn from the network analysis?
Relevant answer
Answer
Where in your research do you employ complex networks and network analysis methods?
I am currently looking at online social movements.
What are your data sources? How big are they?
I can end up with fairly large sets of nodes and edges. Hundreds of thousands of actors when unfiltered, although I usually apply some form of sampling or filtering. For Twitter, events can be pretty good network boundary setters. Always depends on the research question of course.
Which tools do you use for the network analysis process?
I have applied ERGM modelling and standard social network descriptive statistics, although my current favourite for large datasets is using correspondence analysis to analyze bipartite networks.
What did you learn from the network analysis?
This is key. With large datasets learning new things can be difficult - at the best of times you mostly see peculiar relationships that can be hard to explain with any level of validity. Often I find myself puttering around with the data to look for a stable pattern that I can explore more deeply, or use as a model to check against other areas of the data. That said, I learned that there is unexpected power and influenced exercised by groups of Twitter users who are simply adept at finding and curating information on their favourite topic. They can be even more influential than celebrities and impassioned activists in certain contexts.
  • asked a question related to Network Science
Question
73 answers
There is a variety of software packages which provide graph algorithms and network analysis capabilities. As a developer of network analysis algorithms and software, I wonder which tools are most popular with researchers working on real-world data. What are your requirements with respect to usability, scalability etc.? What are the desired features? Are there analysis tasks which you would like to do but are beyond the capabilities of your current tools?
Relevant answer
Answer
Hi
I think DIgSILENT PowerFactory is the best for your simulation
  • asked a question related to Network Science
Question
3 answers
Graph based community detection methods are very effective in explaining the underlying structure of graph but i have not come across any method find optimal number of community similar to clustering methods.
Relevant answer
Answer
I am sorry, I should have made it a bit clear.
Say, I am trying to identify the communities in an unsupervised manner and for that  I am trying to maximize the modularity. Now, I get different number of communities with different nodes even at a single resolution parameter. The question arises which of the communities is the best i.e. is there any statistical criteria which can lead me to find that number? 
Moreover, the choice of resolution parameter itself is a question mark.
  • asked a question related to Network Science
Question
4 answers
In SIS/SIRS model the network components which are infected are assumed to recover and go to susceptible state based immediately or after some time based on immunity of the virus.
While in SIR model, once recovered, the nodes are assumed to have become immune to the same disease and no longer participates in the spread of epidemic.
I was just wondering which model describes the computer virus epidemic more accurately.
Relevant answer
Answer
Hi Manikant,
The immunity is not being provided by the adversary, but by the computer owner himself (or the network admin). If it helps, imagine the network admin's spreading "patches", while the adversary is spreading the virus (with each having multiple versions). At any point of time, a computer node is "safe" (susceptible) if its patch version is later than the virus version it encounters, else it gets infected. So you'll have a SIS dynamic in that case, with transitions happening in both directions.
Of course, I am assuming R-> recovered (and hence a computer node gaining immunity to the current version of the virus). If the virus is 'fatal', then the node gets removed, and you can only have SIR dynamics.
Srini
  • asked a question related to Network Science
Question
6 answers
We know very well that a communication network is always assumed to be or also exhibits the nature of Scale Free Networks. But is it possible that a communication network can be framed as ER Network or a Random Network?
Links to any papers or thesis that model communication networks as ER or Random model would help too.
Relevant answer
Answer
Some wireless networks (sensor networks, ad hoc networks) are best modeled as random geometric graphs, with a Poisson point process as the underlying node distribution and edges depending on the distances between nodes, deterministically or probabilistically. In most cases, the node degrees will be Poisson, but they won't be independent due to the geometric nature of the graph.
  • asked a question related to Network Science
Question
4 answers
Epidemic in Networks
Looking for some of the best papers or thesis to go through.
Relevant answer
Answer
While some of the references below are dated, they should provide you with a good place to start and traditional well established model from which you can start looking for improvements and variations in more recent publications.
Wieland, S., Aquino, T., & Nunes, A. (2012). The structure of coevolving infection networks. EPL (Europhysics Letters), 97(1), 18003.
Kenah, E., & Robins, J. M. (2007). Second look at the spread of epidemics on networks. Physical Review E, 76(3), 036113.
Tao, Z., Zhongqian, F., & Binghong, W. (2006). Epidemic dynamics on complex networks. Progress in Natural Science, 16(5), 452-457.
Saramäki, J., & Kaski, K. (2005). Modelling development of epidemics with dynamic small-world networks. Journal of Theoretical Biology, 234(3), 413-421.
Hethcote, H. W. (2000). The mathematics of infectious diseases. SIAM review, 42(4), 599-653.
Korobeinikov, A. (2004). Global properties of basic virus dynamics models. Bulletin of Mathematical Biology, 66(4), 879-883.
Day, T., & Proulx, S. R. (2004). A general theory for the evolutionary dynamics of virulence. The American Naturalist, 163(4), E40-E63.
May, R. M., & Lloyd, A. L. (2001). Infection dynamics on scale-free networks. Physical Review E, 64(6), 066112.
  • asked a question related to Network Science
Question
11 answers
The human connectome is a comprehensive map of the neural connections in the brain - in other words, a graph. Coming from a background in graph algorithm development and network analysis, the field of connectome analysis seems to me a very interesting application domain. However, it is a domain I am just beginning to understand. Therefore I hope to get some feedback from both neuro- and computer scientists, starting with the following questions:
- It is my understanding that at the neural scale, the connectome is a graph of more than 10^10 nodes and 10^14 edges. If it could be comprehensively mapped at this scale - which i believe it cannot at this point due to a lack of imaging technology - would it be in the range of current computing capabilities to analyse such a network?
- Has the connectome been mapped at coarser scales? If yes, what graph sizes are we talking about?
- Are standard measures from network analysis (such as degree distribution, diameter, clustering coefficients, centrality, communities) relevant for connectome analysis? What are interpretations of such measures?
- What are other structures of interest in the connectome that could be revealed by graph algorithms? Is there a need for domain-specific algorithms to discover brain-specific graph structures?
- Are there publicly available datasets that represent the connectome as a graph?
Relevant answer
Answer
Hi Christian - very good questions (which I stumbled upon quite by accident), and YES, connectome maps at coarser (regional) scales have indeed been compiled (many are publicly available) and have been studied extensively with graph measures and tools. Of particular interest are measures that reveal community structure and hubs - if you email me directly (osporns@indiana.edu) I can send you references as well as data sets that you might find helpful. All best -- Olaf
  • asked a question related to Network Science
Question
3 answers
As long as the air carries electrical signals, why not take advantage of these electrical signals to charge mobile devices, "charging in case of emergency," for example.
Relevant answer
Answer
I guess you must have heard about "witricity" or wireless electricity. But the problem with it is, it can't be deployed everywhere. It needs a proper set up. Check out this Wiki link-
and this one also:
  • asked a question related to Network Science
Question
4 answers
What performance metrics are to be measured for the new MAC protocol for wireless sensor networks with multi-rate sensor nodes?
Especially I focus on proportional fairness between nodes.
Relevant answer
Answer
Thanks a lot
What about throughput?
  • asked a question related to Network Science
Question
3 answers
In a directed network of agents passing knowledge to each other how fast does information in a node grow with the number of edges incident on it? Linearly or exponentially with the number of edges? If on one hand the knowledge of the node agent seems to grow with the sum of the knowledge shared through the incoming edges, on the other hand each item of information shared through one edge might recombine with each item of other edges incoming information, to form new items of information. For example if agent A tells me that there is a traffic jam near the shopping center and agent B tells me that today morning there are big sales there, I get to acquire a third peace of information from combining what agent A and Agent B told me, which is that the traffic jam is caused by the sales and won't stop until the sails are over. Any ideas on which model better suits reality? Sorry if this sounds as a rather naïve question, but since it is related, but not central to my research, I did not get to do a literature search on it.
Relevant answer
Answer
a while ago, i wrote some papers on computational semiotics that are at peripherally relevant. The ``Get Stuck'' Theorems show (1) that no matter what finite representational mechanism you use, if you are trying to represent more and more complex phenomena, you will get stuck (the size of the description exceeds your computational capacity to compute with it), and second, that even if humans are adding to a base of knowledge, eventually the knowledge will become too cumbersome to change, mainly because there will be too many interconnections
reference:
Cognitive Technology: Instruments of Mind: 4th International Conference, CT ...
edited by Meurig Beynon, Chrystopher L. Nehaniv, Kerstin Dautenhahn, Springer LNAI 2117 (2001)
the relevance is that a faster increase in knowledge can lead to getting stuck sooner, so it isn't always the most desirable behavior
  • asked a question related to Network Science
Question
9 answers
Can anyone recommend a site for documenting my research online so that it will be visible to all viewers?
Relevant answer
Answer
If Researchgate is the best then the others must be bl**dy awful. As far as I can work out it's impossible to see a list without the abstracts without turning them off INDIVIDUALLY! I know my publication list is missing about fifty entries, but it's impossible to view it to see which ones. Totally user-unfriendly.
  • asked a question related to Network Science
Question
4 answers
I want to know whether there an online tool exists other than a simulator to evaluate and analyze the realtime performance of a social network?
Relevant answer
Answer
There is a nice compilation of SNA tools here: http://en.wikipedia.org/wiki/Social_network_analysis_software
Just search for "browser" - to get the online tools ( I think there are 4 mentioned in the list)
  • asked a question related to Network Science
Question
4 answers
enterprise architecture can itself be made agile by optimising the organisation it represents, to make it agile. Businessmen however, expect something from this process. ROI? organisational effectiveness? resilinece to change?
Relevant answer
Answer
Enterprise Architecture, to my mind, is the re-engineering of the enterprise in a way similar to that of sculptor chiseling out a form, say human form, and providing stability as well as grandeur. If we compare enterprise with human form, then in case of enterprise, Head is leadership, Body is organization, Skeletal structure is organization structure, Heart is agility, Soul is organization culture, Two hands are supplier and customer, Two supporting legs are employees and organization system. While heart keeps the human body alive and moving, the agility provides organization the adaptability and ability to quick decision making in response to the changing environment. On the other side, organization architecture enables to make necessary changes in the organizational system for sustainability and growth in a changing environment. Agile organizations are flexible, innovative and profitable. Thus agility and enterprise architecture are complementary.
  • asked a question related to Network Science
Question
4 answers
Using GEO data and network inference methods (GRN), I have got relationships between genes from each GRN method (example, GENIE3, CLR, LASSO). For specific genes which I am interested, I have got also the network from literature (text mining based.....Ariadne, Ingenuity, STRING). As all these relationships are purely predictions, how to effectively combine/overlap these predictions (to make an ensemble network) so that I don't loose novel relationship?
GEOdata: Mouse macrophage cell line RAW264.7 with LPS stimulation
Relevant answer
Answer
If you are not concerned with edge weighting, as Jonathan is describing, which would provide relative levels of confidence in each edge, but are simply interested in compiling all possible edges as a predictive network, then you could use Cytoscape to assemble them. Other than the fact that you need to have your data formatted correctly, it should be pretty straightforward to compile. You'd need to carry out some filtering of duplicate interactions in a spreadsheet prior to Cytoscape upload, unless you're not bothered by multiple edges between node pairs, but it should be fairly easy to do. Once it's in Cytoscape, there are many apps/tools that can be helpful to interrogate the final network. The one I'm most familiar with is BiNGO, which allows ontological analysis of your network. Of course, in the end, these are all simply predictions, so it is best to consider the data as a predictive tool, which in turn allows you to generate testable hypotheses that are best assessed experimentally.
  • asked a question related to Network Science
Question
1 answer
Cryptography require?
Relevant answer
Answer
Please be more precise with your question, e.g. which algorithms you are talking about for which applicatio, why you need to know data rates and why you think there must be some "requirement" for those rates.
I strongly advise you to have a read on Daniel J. Bernstein's document about the era of post-quantum cryptography: http://www.pqcrypto.org/www.springer.com/cda/content/document/cda_downloaddocument/9783540887010-c1.pdf
Please be sure to fully understand Chapter 4.
If I had to guess what you mean by your question, I would quote Bernstein:
A modern stream cipher can run on any commonly available CPU, and
generates gigabytes of stream per second on a $200 CPU. Quantum cryptography generates kilobytes of stream per second on special hardware
costing $50000.
So it is not about what kind of data rates quantum cryptography "requires". Modern ciphers also do not have a data rate requirement, you just make use of the data rate available to you through your hard- and software.
It will be more about how to build inexpensive hardware for quantum cryptography in the future to reach high data rates.
  • asked a question related to Network Science
Question
1 answer
How can we predict future network attack ina network?
Relevant answer
Answer
Using Markov chain Modeling
  • asked a question related to Network Science
Question
3 answers
I'm searching for any dataset that includes information of species interactions measured at different sites and times (so that spatial & temporal covariance of species is provided). The idea is to test whether or not, and to what extent, food web properties can be predicted from 'easy-to-collect' data (species presence, absence etc.).
Relevant answer
Answer
Hello
I have 3 sets of island data that might be of some help.
A.
It is full soil fauna (excluding protists) data set + plants on 22 islands (727 species of animals +355 plants). Animals are divided into trophy categories. The first paper is already published here: http://www.sekj.org/PDF/anz49-free/anz49-161i.pdf
There are also important defects of the data
- Sampling effort unknown: main idea of project (year 1973-1983) was creation of faunistic list + showing differences between habitats. There was diverse sampling effort, and that effort is unknown now. The largest island where station was based is studied very well and it might serve as species pool information.
- Abundance of species is unknown
- there is no information on times series
B.
There are also data from my studies on three lake islands archipelagoes. There are data sets for
1. carabids (ca 110 species, ca 50 islands) ; species divided into 3-4 dispersal categories,
2. spiders (ca 180 species, ca 23 islands) we divided species into dispersal categories,
3. plants (ca 300 species, ca 50 islands) we have seed size and what is rare for plants abundance (not Braunblanqet cover),
4. Heteropterans (ca 60species, ca 23 islands) data are for island level not fro trap level,
5. Small Mammals (ca 7 species and only 300 individuals, ca 23 islands) data set is still not in excel format.
C. Stable isotope (C and N) data on 59 carabid species (35 species with decent sample size) collected on 20 islands
You are welcome to play with that data
  • asked a question related to Network Science
Question
7 answers
I want to ask a General question. Many students did not know how to work with protocol, how to see the protocol, how to open it, how to improve it. Please tell me how we can work with protocols.
Relevant answer
Answer
What exactly is your goal? Do you want to explain specific protocols? Do you want to explain protocol design? Do you want to show real-life protocols as they run, or do you want to perform an efficiency analysis?
Depending on those, I think you have several choices, also depending on experience level and course goals:
- network sniffers like tcpdump or wireshark, as already noted by Bruno; great for explaining practical usage (and sometimes security)
- a black/whiteboard; this is a great way to explain the inner workings of a protocol. You can explain the reasoning behind each step as you build up the protocol, and it allows you to show construction too.
- a network simulator; useful for experimenting and rapid prototyping of applications that require specific (expensive) devices (such as vehicular networks).
- implementation task; a good way to have students become intemately familiar with protocols. I wouldn't recommend using RFCs as a basis, though; they are often a very dry read and (thus) not very motivating to students. Implementation is, however, the most practical way to get students familiar with something.
I hope this helps.
  • asked a question related to Network Science
Question
10 answers
Is someone knows where to find datasets of networks with known communities (that's the important point), in order to have reference clusters to validate/invalidate community detection algorithms ?
Relevant answer
Answer
@ Vincent - "known communities" is very tricky, because with different algorithms you get different communities. There aren't any "known" communities since you can assemble the players according to various different characters. Each character might end up with different significant communities. So there isn’t any “Right” answer.
In case you find in the net a dataset with "known" communities clusters - you must know which algorithm, had detected these communities and then you must understand the algorithm, its purpose and rationale And does it suits your research question?
Did I make myself clear?
  • asked a question related to Network Science
Question
19 answers
Can anybody recomend me a book about networks in general?
Thanks a lot
Relevant answer
Answer
Ok, I have done my PhD on the social networks of teachers.
If you want to know more about the analysis of social networks - read Jacob Moreno (1932/1934/1954)
De Nooy, W., Mrvar, A. & Batageli, V. (2006). Exploratory social network analysis with Pajek. New York: Cambridge University Press.
Cross, R. & Parker, A. (2004). The hidden power of social networks: understanding how work really gets done in organizations. Boston: Harvard University Press.
There are lots of very good papers available as well.
  • asked a question related to Network Science
Question
2 answers
Connections are in the following manner
Modem>Router>PC
I tried with modem>PC its working but with the router its not working?
Relevant answer
Answer
@carl
Ya that's rit. But that shows only router settings rather wat the settings should done in modem. I tried the settings what shown in website but its not working.If i using modem alone can access through remote.
  • asked a question related to Network Science
Question
1 answer
Can cloud computing bridge the digital divide in Education?
Relevant answer
Answer
I believe it might. I certainly offers the possibility of using affordable network computers at the user interface level.
  • asked a question related to Network Science
Question
2 answers
For 1-27 series ip address mask is 255.0.0.0 then if i enter like above means whether IP will shown in network or not?
Relevant answer
Answer
it will work.this is a classless ip addresss.
In this ip address
network address-10.213.75.0
Can add only 255 number of host
  • asked a question related to Network Science
Question
2 answers
"Analyzing Social Media Networks with NodeXL: Insights from a Connected World", Derek Hansen, Ben Shneiderman, Marc A. Smith , Publisher: Morgan Kaufmann; 1 edition , 2010
-Or any other good tutorial on how to use NodeXL
Thanks!
Relevant answer
Answer
Thank you!
  • asked a question related to Network Science
Question
4 answers
How can I build a virtual network in the groups of virtual machines?
I need to use with the heartbeat
Relevant answer
Answer
Thank you, again.
I don´t understand, how work this tool.
I need a tools to create a similar with a new network interface eth1 on the guest servers.
And so assign that interface with the tool heartbeat. I work in the virtual enviroment.
  • asked a question related to Network Science
Question
4 answers
I'm doing an M.E and my project is implementing digital signature in bluetooth network to provide security. Can I develop this project by doing some apllication using bluetooth network?
Relevant answer
Answer
RSA and DSA are two different algorithms. RSA can be used both for encrypting and signing, while DSA can only be used for signing. I think DSA is considered more secure if you just want to sign stuff, but I'm not sure about that.
Indeed, it is normal behaviour for it to ask for the passphrase every time. The passphrase is the key for the symmetric cipher that the key is encrypted with, so SSH can't decrypt the key for usage if you don't specify the passphrase. You can change the passphrase with "ssh-keygen -p". You can even choose not to have a passphrase, which is convienent for jumping between computers without using a password. Of course, it's bad if someone would get your private key, since they would be able to do the same, but as long as you keep your secret key in a safe place and unreadable for other users than yourself, you're safe
  • asked a question related to Network Science
Question
1 answer
interested in finding more information on use of magnetic propulsion for future space travel, I feel with enough Guaess force and opposing forces one may be able to create such a craft. using megnetic flight may be the answer to space travel and also the use of pulse lazer light. any question or added comments will help me in my reseach.
  • asked a question related to Network Science
Question
4 answers
Dear Members,
Currently I'am working on molecular epidemiology of Cryptococcus gattii, an important human pathogen. this yeast causes life-threatening infection of the pulmonary and central nervous systems in hosts with normal immunity and traditionally has been considered to be restricted geographically to tropical and subtropical climates.
To date, for reasons that are not yet fully understood, C. gattii has acquired the ability to adapt to new climatic and geographic conditions, such as those existing in Canada, where this yeast has unexpectedly emerged as a primary pathogen causing, since 1999, infections mostly in immunocompetent individuals. Moreover, during the past years, several clinical autochthonous cases have also been described of patients who live in Mediterranean countries showing that this fungus is more widespread than was previously thought. However no environmental studies have been conducted to evaluate the diffusion of this yeast in these geographical areas.
To this day, after twenty years of investigations our laboratory isolated this fungus in Reggio Calabria, southern Italy (from Eucalyptus camaldulensis trees) that emphasize the observed global expansion of this pathogenic yeast.
I therefore invite all those who live in the Mediterranean area to start a collaboration with my laboratory to perform an important environmental study about the spread of this yeast.
I believe that this effort to determine the ecology and population dynamics of C. gattii in Europe might detect a different reality than that currently known regarding the epidemiology of this species.
You can simply participate in the study by sending me the environmental samples (Eucalyptus camaldulensis debris, including leaves and barks) from your country or if you are able you can recover this yeast from these samples in your lab by using conventional phenotypic methods and than share the results.
I hope many of you will participate in this study.
Anyone interested can contact me at oromeo@unime.it for further details.
I think it will be a wonderful (and productive) experience working with you.
Best Regards
Orazio Romeo
Relevant answer
Answer
Dear Juan Bosco Guzmán Pérez,
Recently I isolated C. gattii in Italy from Eucalyptus trees and now I decided to start an European project to describe the environmental diffusion of this yeast in the Mediterranean area.
My laboratory also works on clinical Cryptococcus spp. isolates. Currently we are performing molecular studies to describe the incidence of molecular types in southern Italy.
Ciao da Messina, Italia
Orazio
  • asked a question related to Network Science
Question
1 answer
What if you made a computer with eproms, put it together even replacing your own chips in a manufactured laptop, even did a token ring sort of thing with different embedded OSs, not just linux, and with different architectures like CISC, RISC etc., so it passes the "job" like a distributed computing or server model, it's easy like moving to a different memory register, and almost impossible to hack, and use randomization and encryption, and even have a system for writing new OSs on the fly with A.I., and use firmware to update andd "reboot" each eprom? I say, make a wearable computer with eproms and different embedded OSs, A.I. or algorithms that rewrite the OS on the fly, and use different little CPUs like CISC, RISC and those from PDAs like Xscale etc., and pass the threads from CPU/OS to each other like a multithreaded network (token ring or ethernet or a new type or make it morph like a virus), and use encryption, firmware, and that makes the OS nearly impossible to hack and also redundant. It can fit in a phone, in a laptop, wearable computer, Tablet computer, and treat different CPUS like both registers and also computing clusters, depending, or randomized, and have it be sandboxed, honeypotted, and "self aware" as in a tripwire type thing that also does things about the issue, not just "reports it." and have it learn from attacks, so that it doesn't need to rely on "virus definitions." Have each eprom connected to it's own CPU be able to reboot on it's own, and not interrupt the other eproms or whatever. Perhaps also have some on backup or standby and have them be hot swapable etc. Lots of combos you can use and also improvements including capacitors, built in solar cells for recharging, built in GPS, Wimax, WiFi, 3G/4G as well as capability to add an antenna internally and even program your own firmware for future technology and also include packet radio to use internet with HAM radio and much more. Give it infrared capability over long distance, microwave, satellite uplink, human mind interface, fiber optics, weather proof/shock proof, shortwave/long wave transceiving, military/aircraft/police frequency capability via eprom uplink/firmware uplink by flash drive with proper authorization codes needed etc.
This is a quit writing in an apple store, would love this Macbook or even an iPad especially if anyone can give me a job designing stuff like this, I've got other ideas and could use a bit of an education, good food etc and especially shelter, am currently homeless. Hey, I'll work for the military as well or law enforcement. I also have ideas on how to deal with militias, organized crime, spies etc. Thanks!
Relevant answer
Answer
Please look at my posts in my other groups, I've got some decent ideas. :)