Styjun

Why are Star Wars Rebel Alliance ships named after letters from the Latin alphabet?

What is Albrecht Dürer's Perspective Machine drawing style?

Why do we need a voltage divider when we get the same voltage at the output as the input?

Why do my fried eggs start browning very fast?

Is the EU really banning "toxic propellants" in 2020? How is that going to work?

speaker impedence

On the expression " sun-down"

How long should I wait to plug in my refrigerator after unplugging it?

Were there any unmanned expeditions to the moon that returned to Earth prior to Apollo?

Can I shorten this filter, that finds disk sizes over 100G?

HackerRank Implement Queue using two stacks Solution

Is verification of a blockchain computationally cheaper than recreating it?

Export economy of Mars

Feedback diagram

What's the term for a group of people who enjoy literary works?

Can birds evolve without trees?

How do people drown while wearing a life jacket?

Gold Battle KoTH

Went to a big 4 but got fired for underperformance in a year recently - Now every one thinks I'm pro - How to balance expectations?

Deflecting lasers with lightsabers

What is time? Does it flow linearly? If so, how are we sure?

How to structure presentation to avoid getting questions that will be answered later in the presentation?

Return last number in sub-sequences in a list of integers

Skipping same old introductions

Is there an algorithm for detecting odd values relative to set of data

Best algorithm for detecting cycles in a directed graphFastest sort of fixed length 6 int arrayDetecting rare incidents from multivariate time series intervalsAlgorithm to detect overlapping periodsThresholding algorithm - positive and negative threshold?Algorithm: Minimum number of TransanctionsGet all possible combination from an arraylist of arraylist of an objectAdvice on what methodology/data structure/algorithm to useAlgorithm for detecting clusters in peaks in time series signal dataTensorflow Object Detection: Not Predicting using Custom Data

.everyoneloves__top-leaderboard:empty,.everyoneloves__mid-leaderboard:empty,.everyoneloves__bot-mid-leaderboard:empty margin-bottom:0;

0

I want to develop an algorithm that makes an action if it detects that some numbers are odd relative to another array of numbers and each number has a date. Those numbers can vary across the day, but the rate of change of these numbers is not necessarily related.

For example,
The data can be

 [
 number: 200, date: '12:00',
 number: 250, date: '12:02',
 number: 180, date: '12:04',
 number: 500, date: '12:06'
 ]

and the array that I want to test is

 [
 number: 400, date: '12:08',
 number: 50, date: '12:10'
 ]

I gather these data for a defined time interval (the above interval is two minutes)
I want to detect if the data falls as time passes, but it can't be directly measured with previous data as it's not consistent and it can fall and rise but I want it to check on the long term.

My Question is what is the approach that I should pursue? Do I have to train a model for that task? If so, what approach should I implement?
I was thinking of writing some hard coded rules that measures the average and compares the data with a threshold. But it wasn't effective on large sets of data as it is not consistent as I stated.

If you have any helpful resources that can help I would be very thankful.

P.S. The above data is not real.

Thanks in advance.

algorithm machine-learning data-structures artificial-intelligence handler

share|improve this question

asked Mar 27 at 0:26

Amr MohamedAmr Mohamed

1

• 
  
  
  

  
  1
  

  
  
  
  
  
  

  
  
  What is the connection between the two arrays? From your question, I understand that you want to detect if a given number series is approximately decreasing. Where does the second series come into play?
  
  – Nico Schertler
  Mar 27 at 3:23
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  Welcome to SO, which is about specific coding questions and not design or recommendation ones ; your question is way too broad, please do take some time to read How to Ask and What topics can I ask about here?.
  
  – desertnaut
  Mar 27 at 10:06
  

  
  
  
  

add a comment | 

0

I want to develop an algorithm that makes an action if it detects that some numbers are odd relative to another array of numbers and each number has a date. Those numbers can vary across the day, but the rate of change of these numbers is not necessarily related.

For example,
The data can be

 [
 number: 200, date: '12:00',
 number: 250, date: '12:02',
 number: 180, date: '12:04',
 number: 500, date: '12:06'
 ]

and the array that I want to test is

 [
 number: 400, date: '12:08',
 number: 50, date: '12:10'
 ]

I gather these data for a defined time interval (the above interval is two minutes)
I want to detect if the data falls as time passes, but it can't be directly measured with previous data as it's not consistent and it can fall and rise but I want it to check on the long term.

My Question is what is the approach that I should pursue? Do I have to train a model for that task? If so, what approach should I implement?
I was thinking of writing some hard coded rules that measures the average and compares the data with a threshold. But it wasn't effective on large sets of data as it is not consistent as I stated.

If you have any helpful resources that can help I would be very thankful.

P.S. The above data is not real.

Thanks in advance.

algorithm machine-learning data-structures artificial-intelligence handler

share|improve this question

asked Mar 27 at 0:26

Amr MohamedAmr Mohamed

1

• 
  
  
  

  
  1
  

  
  
  
  
  
  

  
  
  What is the connection between the two arrays? From your question, I understand that you want to detect if a given number series is approximately decreasing. Where does the second series come into play?
  
  – Nico Schertler
  Mar 27 at 3:23
  

  
  
  
  


• 
  
  
  

  
  

  
  
  
  
  
  

  
  
  Welcome to SO, which is about specific coding questions and not design or recommendation ones ; your question is way too broad, please do take some time to read How to Ask and What topics can I ask about here?.
  
  – desertnaut
  Mar 27 at 10:06
  

  
  
  
  

add a comment | 

I want to develop an algorithm that makes an action if it detects that some numbers are odd relative to another array of numbers and each number has a date. Those numbers can vary across the day, but the rate of change of these numbers is not necessarily related.

For example,
The data can be

 [
 number: 200, date: '12:00',
 number: 250, date: '12:02',
 number: 180, date: '12:04',
 number: 500, date: '12:06'
 ]

and the array that I want to test is

 [
 number: 400, date: '12:08',
 number: 50, date: '12:10'
 ]

I gather these data for a defined time interval (the above interval is two minutes)
I want to detect if the data falls as time passes, but it can't be directly measured with previous data as it's not consistent and it can fall and rise but I want it to check on the long term.

My Question is what is the approach that I should pursue? Do I have to train a model for that task? If so, what approach should I implement?
I was thinking of writing some hard coded rules that measures the average and compares the data with a threshold. But it wasn't effective on large sets of data as it is not consistent as I stated.

If you have any helpful resources that can help I would be very thankful.

P.S. The above data is not real.

Thanks in advance.

algorithm machine-learning data-structures artificial-intelligence handler

share|improve this question

asked Mar 27 at 0:26

Amr MohamedAmr Mohamed

1

 [
 number: 200, date: '12:00',
 number: 250, date: '12:02',
 number: 180, date: '12:04',
 number: 500, date: '12:06'
 ]

 [
 number: 400, date: '12:08',
 number: 50, date: '12:10'
 ]

share|improve this question

asked Mar 27 at 0:26

Amr MohamedAmr Mohamed

1

share|improve this question

asked Mar 27 at 0:26

Amr MohamedAmr Mohamed

1

asked Mar 27 at 0:26

Amr MohamedAmr Mohamed

1

asked Mar 27 at 0:26

Amr MohamedAmr Mohamed

1

1 Answer
1

active

oldest

votes

0

You want an outlier detection which is only searching for decreases.

I propose to create a kernel which predicts the next value based on the recent ones. See http://nbviewer.jupyter.org/github/SheffieldML/notebook/blob/master/GPy/basic_gp.ipynb for starters. The kernel can give you a prediction as well as a confidence margin. If your actual value is more than a certain distance below the allowed confidence margin of the prediction, you can call that an outlier in the negative direction and react on it: Diagram of a confidence margin

share|improve this answer

answered Mar 27 at 14:26

AlfeAlfe

34.8k1212 gold badges6767 silver badges120120 bronze badges

add a comment | 

Your Answer

StackExchange.ifUsing("editor", function ()
StackExchange.using("externalEditor", function ()
StackExchange.using("snippets", function ()
StackExchange.snippets.init();
);
);
, "code-snippets");

StackExchange.ready(function()
var channelOptions =
tags: "".split(" "),
id: "1"
;
initTagRenderer("".split(" "), "".split(" "), channelOptions);

StackExchange.using("externalEditor", function()
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled)
StackExchange.using("snippets", function()
createEditor();
);

else
createEditor();

);

function createEditor()
StackExchange.prepareEditor(
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader:
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
,
onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
);



);

draft saved

draft discarded

Sign up or log in

StackExchange.ready(function ()
StackExchange.helpers.onClickDraftSave('#login-link');
);

Sign up using Google

Sign up using Facebook

Sign up using Email and Password

Post as a guest

Name

Email

Required, but never shown

StackExchange.ready(
function ()
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fstackoverflow.com%2fquestions%2f55368078%2fis-there-an-algorithm-for-detecting-odd-values-relative-to-set-of-data%23new-answer', 'question_page');

);

Post as a guest

Name

Email

Required, but never shown

0

You want an outlier detection which is only searching for decreases.

I propose to create a kernel which predicts the next value based on the recent ones. See http://nbviewer.jupyter.org/github/SheffieldML/notebook/blob/master/GPy/basic_gp.ipynb for starters. The kernel can give you a prediction as well as a confidence margin. If your actual value is more than a certain distance below the allowed confidence margin of the prediction, you can call that an outlier in the negative direction and react on it: Diagram of a confidence margin

share|improve this answer

answered Mar 27 at 14:26

AlfeAlfe

34.8k1212 gold badges6767 silver badges120120 bronze badges

add a comment | 

0

You want an outlier detection which is only searching for decreases.

I propose to create a kernel which predicts the next value based on the recent ones. See http://nbviewer.jupyter.org/github/SheffieldML/notebook/blob/master/GPy/basic_gp.ipynb for starters. The kernel can give you a prediction as well as a confidence margin. If your actual value is more than a certain distance below the allowed confidence margin of the prediction, you can call that an outlier in the negative direction and react on it: Diagram of a confidence margin

share|improve this answer

answered Mar 27 at 14:26

AlfeAlfe

34.8k1212 gold badges6767 silver badges120120 bronze badges

add a comment | 

You want an outlier detection which is only searching for decreases.

I propose to create a kernel which predicts the next value based on the recent ones. See http://nbviewer.jupyter.org/github/SheffieldML/notebook/blob/master/GPy/basic_gp.ipynb for starters. The kernel can give you a prediction as well as a confidence margin. If your actual value is more than a certain distance below the allowed confidence margin of the prediction, you can call that an outlier in the negative direction and react on it: Diagram of a confidence margin

share|improve this answer

answered Mar 27 at 14:26

AlfeAlfe

34.8k1212 gold badges6767 silver badges120120 bronze badges

share|improve this answer

answered Mar 27 at 14:26

AlfeAlfe

34.8k1212 gold badges6767 silver badges120120 bronze badges

answered Mar 27 at 14:26

AlfeAlfe

34.8k1212 gold badges6767 silver badges120120 bronze badges

answered Mar 27 at 14:26

AlfeAlfe

34.8k1212 gold badges6767 silver badges120120 bronze badges