Neurobot (OP)
Newbie
Offline
Activity: 25
Merit: 0
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
October 20, 2017, 11:29:51 PM |
|
|
|
|
|
CryptoRobert
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
October 20, 2017, 11:39:32 PM |
|
Project seems very interesting, I will check your Neurobot for a few days. Do you foresee any bounties to promote it? Concurrence is big in this field.
|
|
|
|
mr_crypt
Newbie
Offline
Activity: 4
Merit: 0
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
October 20, 2017, 11:45:58 PM |
|
To train the neural network you use historical data ?
|
|
|
|
Neurobot (OP)
Newbie
Offline
Activity: 25
Merit: 0
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
October 20, 2017, 11:53:37 PM |
|
To train the neural network you use historical data ?
We have a training sample is collected on the basis of technical analysis.
|
|
|
|
achilles3791
Newbie
Offline
Activity: 52
Merit: 0
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
October 21, 2017, 04:16:46 PM |
|
Can you provide a brief explanation of the forecasting algorithm?
|
|
|
|
Neurobot (OP)
Newbie
Offline
Activity: 25
Merit: 0
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
October 21, 2017, 04:35:06 PM |
|
To train the neural network you use historical data ?
We have a training sample is collected on the basis of technical analysis. do you develop your own machine learning algorithm or use well known? and how long does it take to train with refreshed set of historical data? Yes, we develop our own algorithm on the basis of the LSTM network and gradient boosting. Historical data are not used!! We manually collected the sample on the basis of technical analysis patterns. We complemented signals are MACD. Next year we will launch a fundamental analysis. News, tweets, posts will be analyzed by the algorithm of the tone of the text and of the received signal will be used in the algorithm technical analysis. Also, we'll extend the algorithm of the technical analysis will be added to the Elliott Wave. Training takes place on a powerful GPU Titan Xp, and takes approximately 40 minutes. We use the library Keras+Theano for gradient boosting is used XGBoost.
|
|
|
|
Neurobot (OP)
Newbie
Offline
Activity: 25
Merit: 0
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
October 21, 2017, 10:23:10 PM |
|
Can you provide a brief explanation of the forecasting algorithm?
NeuroBot’s algorithm is based on LSTM-network (Long short-term memory) coupled with Gradient Boosting - another machine learning technique. Thanks to the abilities of neural networks, NeuroBot is constantly learning to better correspond to the state of the market, to make more accurate forecasts and improve the quality of the resulting work. Our algorithms are unique. We do not disclose the details of the work of our neural network, this is a commercial secret ![Wink](https://bitcointalk.org/Smileys/default/wink.gif)
|
|
|
|
|
Dump3er
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
October 23, 2017, 09:59:54 AM |
|
Does that even make sense? How large is the data set that you base your conclusions on with the 70% to 99%? And which currencies does that cover? How does your algorithm adjust when literally anybody would start using your system?
we test project since may and apprx. accuracy for Alpha free version around 75-82 percent depends on chart and currency . Please visit forecats page to check that yourself https://neurobot.trading/forecasts. BTC, ETH, LTC, XRP, DASH, XMR. If you want to ask real users about there expirience join in our telegram group https://t.me/joinchat/CNzks0L-X3IDzHgKIDHU9Q . in January Beta versiont starts all forecasts will be chared with NEB Token, that means that information will be closed from the general public . Algorism is learning algorith, it ll be beter in time, especially when we implement neural market news analysys in the algorithm. Building strong neural networks requires an incredible amount of expertise. Otherwise you end up building a system following the good old "garbage in - garbage out" theory. I see a few problems with projects claiming enormous rates of accuracy based on rather small data sets. And I would be really interested what happens if people use your system and everybody follows the predictions. How is that going to influence the accuracy? Is it making it better? Will we reach some sort of equilibrium (which I don't think because people will keep looking for the alpha)? Difficult question and I don't really know the answer.
|
SOMETIMES YOU WIN, SOMETIMES YOU LEARN!
|
|
|
Factor0008
Newbie
Offline
Activity: 57
Merit: 0
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
October 25, 2017, 12:01:51 AM |
|
Hi, i've studied technical analysis for 7 years already, also i could not see any bounty page. There's any work you guys would like to bounty on it? Can be either graphical analysis, as also translation of whitepaper, ann thread from english to portuguese.
Let me know!
|
|
|
|
Neurobot (OP)
Newbie
Offline
Activity: 25
Merit: 0
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
October 25, 2017, 01:07:35 AM |
|
Does that even make sense? How large is the data set that you base your conclusions on with the 70% to 99%? And which currencies does that cover? How does your algorithm adjust when literally anybody would start using your system?
we test project since may and apprx. accuracy for Alpha free version around 75-82 percent depends on chart and currency . Please visit forecats page to check that yourself https://neurobot.trading/forecasts. BTC, ETH, LTC, XRP, DASH, XMR. If you want to ask real users about there expirience join in our telegram group https://t.me/joinchat/CNzks0L-X3IDzHgKIDHU9Q . in January Beta versiont starts all forecasts will be chared with NEB Token, that means that information will be closed from the general public . Algorism is learning algorith, it ll be beter in time, especially when we implement neural market news analysys in the algorithm. Building strong neural networks requires an incredible amount of expertise. Otherwise you end up building a system following the good old "garbage in - garbage out" theory. I see a few problems with projects claiming enormous rates of accuracy based on rather small data sets. And I would be really interested what happens if people use your system and everybody follows the predictions. How is that going to influence the accuracy? Is it making it better? Will we reach some sort of equilibrium (which I don't think because people will keep looking for the alpha)? Difficult question and I don't really know the answer. We have an experienced team, and if everyone will use the predictions of the NeuroBot and buy/sell the crypto based on them, the accuracy will only improve.
|
|
|
|
Dump3er
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
October 26, 2017, 12:37:45 AM |
|
Does that even make sense? How large is the data set that you base your conclusions on with the 70% to 99%? And which currencies does that cover? How does your algorithm adjust when literally anybody would start using your system?
we test project since may and apprx. accuracy for Alpha free version around 75-82 percent depends on chart and currency . Please visit forecats page to check that yourself https://neurobot.trading/forecasts. BTC, ETH, LTC, XRP, DASH, XMR. If you want to ask real users about there expirience join in our telegram group https://t.me/joinchat/CNzks0L-X3IDzHgKIDHU9Q . in January Beta versiont starts all forecasts will be chared with NEB Token, that means that information will be closed from the general public . Algorism is learning algorith, it ll be beter in time, especially when we implement neural market news analysys in the algorithm. Building strong neural networks requires an incredible amount of expertise. Otherwise you end up building a system following the good old "garbage in - garbage out" theory. I see a few problems with projects claiming enormous rates of accuracy based on rather small data sets. And I would be really interested what happens if people use your system and everybody follows the predictions. How is that going to influence the accuracy? Is it making it better? Will we reach some sort of equilibrium (which I don't think because people will keep looking for the alpha)? Difficult question and I don't really know the answer. We have an experienced team, and if everyone will use the predictions of the NeuroBot and buy/sell the crypto based on them, the accuracy will only improve. For now that is just an assumption you make. I am not saying you are wrong, and the reason is that I simply don't now. I find it an interesting experiment to see a market where the majority follows artificially calculated advice. I am certain there will be a lot of things in the future we can't even think of yet. AI will play a role of course.
|
SOMETIMES YOU WIN, SOMETIMES YOU LEARN!
|
|
|
Neurobot (OP)
Newbie
Offline
Activity: 25
Merit: 0
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
October 26, 2017, 09:26:05 PM |
|
Hi, i've studied technical analysis for 7 years already, also i could not see any bounty page. There's any work you guys would like to bounty on it? Can be either graphical analysis, as also translation of whitepaper, ann thread from english to portuguese.
Let me know!
Soon we will post the bounty program
|
|
|
|
Neurobot (OP)
Newbie
Offline
Activity: 25
Merit: 0
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
October 26, 2017, 09:39:30 PM |
|
To train the neural network you use historical data ?
We have a training sample is collected on the basis of technical analysis. do you develop your own machine learning algorithm or use well known? and how long does it take to train with refreshed set of historical data? Yes, we develop our own algorithm on the basis of the LSTM network and gradient boosting. Historical data are not used!! We manually collected the sample on the basis of technical analysis patterns. We complemented signals are MACD. Next year we will launch a fundamental analysis. News, tweets, posts will be analyzed by the algorithm of the tone of the text and of the received signal will be used in the algorithm technical analysis. Also, we'll extend the algorithm of the technical analysis will be added to the Elliott Wave. Training takes place on a powerful GPU Titan Xp, and takes approximately 40 minutes. We use the library Keras+Theano for gradient boosting is used XGBoost. Thanks, looks sound. Probably will not consider myself as investor but rather as a customer to try your bot at my trading routine. From 1 January we are introducing a paid subscription. Tokens NEB can be used to pay for the forecasts. You better invest in our project now and get a bonus of up to 60% on pre-sale.
|
|
|
|
|
Neurobot (OP)
Newbie
Offline
Activity: 25
Merit: 0
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
October 27, 2017, 12:33:52 PM |
|
|
|
|
|
Neurobot (OP)
Newbie
Offline
Activity: 25
Merit: 0
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
October 27, 2017, 05:21:21 PM |
|
Guys, we remind that the subscription to forecasts from January 1 will be paid. The cost will start from $1200 per year or 16000 NEB. When you purchase 13334 NEB, you get a 20% bonus, that's 2666 NEB. All together 16000 NEB - this is enough to pay for a subscription for 1 year. Do not miss a great offer! Invest in Neurobot! https://neurobot.trading/presale
|
|
|
|
PokerDiceMan
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
October 27, 2017, 06:40:11 PM |
|
youre project only bot trading, why must create coin and open ico selling why not direct selling bot
|
|
|
|
Dump3er
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
November 01, 2017, 12:09:33 PM |
|
youre project only bot trading, why must create coin and open ico selling why not direct selling bot
This is much more than bot trading You need data of highest integrity and blockchain is able to deliver the integrity of the data that is required.
|
SOMETIMES YOU WIN, SOMETIMES YOU LEARN!
|
|
|
Ale88
Legendary
Offline
Activity: 2492
Merit: 2737
|
![](https://bitcointalk.org/Themes/custom1/images/post/xx.gif) |
November 24, 2017, 04:45:40 PM |
|
Guys, we remind that the subscription to forecasts from January 1 will be paid. The cost will start from $1200 per year or 16000 NEB. When you purchase 13334 NEB, you get a 20% bonus, that's 2666 NEB. All together 16000 NEB - this is enough to pay for a subscription for 1 year. Do not miss a great offer! Invest in Neurobot! https://neurobot.trading/presaleWhat happened with the presale? Why has it been canceled?
|
|
|
|
|