For normalisation of the matrix I think you are using the linear normalisation (sum) in this case, then for the cost criteria you must take the reciprocal of the values as explained by you in your normalisation video....
Indeed, the normalisation was a mistake. However, as I consider cost as a non-beneficial criterion, the final weights are still similar to the result when cost was consider as a beneficial one. Here is my result, considering cost is non-beneficial criterion, its weight is 0.0646, and storage, camera and looks are 0.3781, 0.2874 and 0.2688 respectively. As others mentioned, cost should be important, but based on this method, it seems not the case.
Thank you my friend. I have a question, will it be valid if I take some portion of my dataset (say 60%) to calculate the weights, then use the remaining portion (40%) for the application?
Dear Sir, Kindly give one example with factors, like some barriers or some attributes or factors, because i dont need to select any mobile or any product i want to know how simply methods can be used to give rank
Can we take these final weight or calculation as ranks ? or taken these as a rank if we want to rank the different alternative and criteria or we have to Use different MCDM techniques for ranking after finding weight through entropy ?? Please reply as soon as possible it's urgent
thak you so much sir ... i realy appreciate that .. does any one here have a good idea about "entropy based weights" or can tell were can i find good articles to better understanding please ?
For this condition prominent numeric values are there(if storage means 32gb like that) so we can find it by adding it. But for instance we are comparing without numeric values. can i assume the weight and follow the procedure?
Hi, I have a question please: What if some values are randomly missing? E.g. in row one, the "storage" value is missing for Mobile 1; in row three, "camera" value is missing for Mobile 3. In that case, how to calculate the h, e and d please?
@@manojmathew5287 Hi Manoj, Thanks for getting back. But in my case, I can't do that. Because every alternative come with incomplete data: each "row" (alternative) has 5 attributes, but only about half of the attributes have a value (that's the nature of the data). Let's assume the missing data is completely random. I'm thinking of the following options: 1. Replacing the missing values with the mean value, for each attribute. 2. Tweaking the Entropy Method calculation e.g. apply different values of "m" (number of alternatives) accordingly. Which I'm nervous about as I don't know this method well enough. What's your advise please?
Sir if I take 5 year data of each mobile phone to find the best of them then what I have to do is first of adding 5 year data of each mobile phone to make to make a composite observation as you have shown above then computing their weights, am I right sir?
Sir, can entropy method be used for determining the weights when the values for attributes for each criteria are linguistic, like in the AHP example you had shown ?
The weightages given by entropy method is it valid enough sir??....I don't think so, because in this problem the weightages of the price is less than all other criterias even less than looks!!!....my goodness....but in reality it must be given as the highest priority I think..... while purchasing a Mobile, I hope price must be the most important criteria that customers look into. After fixing a budget all the buyers choose any Mobile model and concentrate on the specifications I think.....this method is not valid enough I hope.....
yes, it is, acc to me he made mistake in normalization part, for non-beneficial criteria the formula for calculation is different. Watch his normalization video under linear normalization(sum) you find the correct formula to calculate non-beneficial criteria.
Can I use this method for assigning weightage in the problem of TOPSIS also? Technique for Order Preference by Similarity to Ideal Solution (TOPSIS)-Click here
![BLACK BAG - Official Trailer [HD] - Only in Theaters March 14](http://i.ytimg.com/vi/Du0Xp8WX_7I/mqdefault.jpg)
Thank you, teacher. Highly appreciated. Easy to understand, straighforward and save my time a lot.
Thanks a lot for your contributions. You are a savior. Keep doing great stuff like this.
Thank you so much please continue doing this type of videos.
Thank you sir.. Its very difficult to understand concepts by reading from literature..
Thank you so much for this content, you made it easy for us
For normalisation of the matrix I think you are using the linear normalisation (sum) in this case, then for the cost criteria you must take the reciprocal of the values as explained by you in your normalisation video....
Indeed, the normalisation was a mistake. However, as I consider cost as a non-beneficial criterion, the final weights are still similar to the result when cost was consider as a beneficial one. Here is my result, considering cost is non-beneficial criterion, its weight is 0.0646, and storage, camera and looks are 0.3781, 0.2874 and 0.2688 respectively. As others mentioned, cost should be important, but based on this method, it seems not the case.
Thank you so much this helped greatly for excel calculations!
in this method are we needed the expertes to give us the numerical numbers? how we give the numbers from experts
Thank you very much to explain the entropy method
Thanks for your video! It helps a lot
Well done Manoj sab
Congratultion Manoj sir
very simple explanation. impressive 👍
Thank you my friend. I have a question, will it be valid if I take some portion of my dataset (say 60%) to calculate the weights, then use the remaining portion (40%) for the application?
Many thanks sir. Your video's are exceptional!
Hi how r u can you explain the SWARA method and Coopers technique . with my greet to u..
Step 1 is wrong....Normalized value for Non Benefical Criteria should be reciprocal of Xij values....ie (1/Xij )/Sum of (1/Xij) values
What? who told it?
Please make videos on latest OBJECTIVE WEIGHTS determination techniques.
Hi Manoj. İnitially ı want to thans to you for your example. Can you give example and explain abaout the fuzzy crıtıc method and its calculating
Dear Sir, Kindly give one example with factors, like some barriers or some attributes or factors, because i dont need to select any mobile or any product i want to know how simply methods can be used to give rank
Thanks for your explanation, can you give us the implementation with R Studio
Can we take these final weight or calculation as ranks ? or taken these as a rank if we want to rank the different alternative and criteria or we have to Use different MCDM techniques for ranking after finding weight through entropy ?? Please reply as soon as possible it's urgent
thak you so much sir ... i realy appreciate that ..
does any one here have a good idea about "entropy based weights" or can tell were can i find good articles to better understanding please ?
Thank you sir for such a useful video. If we also want to calculate weighing factor for sub-criteria, so how can we incorporate SE method?
Your video is very helpful sir. Can you please make a video of Improved Entropy method?
Thanks very much for these videos
Can you explain same example by fuzzy topsis method?
Good one.. 👍👍
For this condition prominent numeric values are there(if storage means 32gb like that) so we can find it by adding it. But for instance we are comparing without numeric values. can i assume the weight and follow the procedure?
Hi, I have a question please: What if some values are randomly missing? E.g. in row one, the "storage" value is missing for Mobile 1; in row three, "camera" value is missing for Mobile 3. In that case, how to calculate the h, e and d please?
Remove those alternatives from the decision matrix. This is called data cleaning process.
@@manojmathew5287 Hi Manoj, Thanks for getting back. But in my case, I can't do that. Because every alternative come with incomplete data: each "row" (alternative) has 5 attributes, but only about half of the attributes have a value (that's the nature of the data). Let's assume the missing data is completely random. I'm thinking of the following options:
1. Replacing the missing values with the mean value, for each attribute.
2. Tweaking the Entropy Method calculation e.g. apply different values of "m" (number of alternatives) accordingly. Which I'm nervous about as I don't know this method well enough.
What's your advise please?
Sir if I take 5 year data of each mobile phone to find the best of them then what I have to do is first of adding 5 year data of each mobile phone to make to make a composite observation as you have shown above then computing their weights, am I right sir?
can give us the example to criteria and how we result the weight of every criteria
Sir, can entropy method be used for determining the weights when the values for attributes for each criteria are linguistic, like in the AHP example you had shown ?
Is the entropy weight supposed to sum to 1?
The weightages given by entropy method is it valid enough sir??....I don't think so, because in this problem the weightages of the price is less than all other criterias even less than looks!!!....my goodness....but in reality it must be given as the highest priority I think..... while purchasing a Mobile, I hope price must be the most important criteria that customers look into. After fixing a budget all the buyers choose any Mobile model and concentrate on the specifications I think.....this method is not valid enough I hope.....
yes, it is, acc to me he made mistake in normalization part, for non-beneficial criteria the formula for calculation is different. Watch his normalization video under linear normalization(sum) you find the correct formula to calculate non-beneficial criteria.
Yah I noticed that.... check my other comment, I already mentioned that in the comment box.....
do u have an excel sheet?
Any one having MATLAB coding for different aggregation operators and decision making technique in fuzzy, intuitionistic, cubic and neutrosophic sets??
thanks
I think in the last step weightage FOR Cost is wrong
U buy phones consider storage space first this is make sense buy I don't know why the price weighting so low
Can I use this method for assigning weightage in the problem of TOPSIS also?
Technique for Order Preference by Similarity to Ideal Solution (TOPSIS)-Click here
Yes you can use the same methodology for evaluating the weights in TOPSIS
Can you explain same example by fuzzy topsis method?