Which between a DL autoencoder, Twitter STL+ESD and Numenta HBM is more state-of-the-art for anomaly...












0












$begingroup$


I aim to develop a generic algorithm to detect abnormalities in time series (ideally accepting multivariate series, and the presence of some noise). The aspects which I am considering for comparing these, are mainly:




  1. Seasonal variations

  2. Amplitude variations on periods.

  3. Period length variations.

  4. Delays (missing data) between periods

  5. Single points variations


I have read about many algorithms, and the ones which I found more state of the art, are these:




  1. (year?) Using Deep learning autoencoders & decide period lengths via Fast Fourier Transform or ACF. ----- Article ----- Implementation

  2. (Apr 2017) Twitter STL + ESD algorithm. ----- Article ----- Implementation.

  3. (Nov 2017) Numenta HTM. Article. ---- Implementation


Numenta also provides in this repository a benchmark between algorithms for anomaly detection. However, (1) I do not know if this benchmark englobes all main methods for this task, and (2) I am not sure if the way they did the benchmark include all the 5 abnormalities I mentioned.



Also, I saw there are more algorithms, like the ones developed here, and I have seen the classification of anomaly detection in basic, statistical and probabilistic ways here, but there is way too much information around for me to figure out my next question.



My question is:



Which algorithm / minimal algorithms combination will [best] detect the kinds of abnormalities aforementioned and why?










share|improve this question











$endgroup$

















    0












    $begingroup$


    I aim to develop a generic algorithm to detect abnormalities in time series (ideally accepting multivariate series, and the presence of some noise). The aspects which I am considering for comparing these, are mainly:




    1. Seasonal variations

    2. Amplitude variations on periods.

    3. Period length variations.

    4. Delays (missing data) between periods

    5. Single points variations


    I have read about many algorithms, and the ones which I found more state of the art, are these:




    1. (year?) Using Deep learning autoencoders & decide period lengths via Fast Fourier Transform or ACF. ----- Article ----- Implementation

    2. (Apr 2017) Twitter STL + ESD algorithm. ----- Article ----- Implementation.

    3. (Nov 2017) Numenta HTM. Article. ---- Implementation


    Numenta also provides in this repository a benchmark between algorithms for anomaly detection. However, (1) I do not know if this benchmark englobes all main methods for this task, and (2) I am not sure if the way they did the benchmark include all the 5 abnormalities I mentioned.



    Also, I saw there are more algorithms, like the ones developed here, and I have seen the classification of anomaly detection in basic, statistical and probabilistic ways here, but there is way too much information around for me to figure out my next question.



    My question is:



    Which algorithm / minimal algorithms combination will [best] detect the kinds of abnormalities aforementioned and why?










    share|improve this question











    $endgroup$















      0












      0








      0


      1



      $begingroup$


      I aim to develop a generic algorithm to detect abnormalities in time series (ideally accepting multivariate series, and the presence of some noise). The aspects which I am considering for comparing these, are mainly:




      1. Seasonal variations

      2. Amplitude variations on periods.

      3. Period length variations.

      4. Delays (missing data) between periods

      5. Single points variations


      I have read about many algorithms, and the ones which I found more state of the art, are these:




      1. (year?) Using Deep learning autoencoders & decide period lengths via Fast Fourier Transform or ACF. ----- Article ----- Implementation

      2. (Apr 2017) Twitter STL + ESD algorithm. ----- Article ----- Implementation.

      3. (Nov 2017) Numenta HTM. Article. ---- Implementation


      Numenta also provides in this repository a benchmark between algorithms for anomaly detection. However, (1) I do not know if this benchmark englobes all main methods for this task, and (2) I am not sure if the way they did the benchmark include all the 5 abnormalities I mentioned.



      Also, I saw there are more algorithms, like the ones developed here, and I have seen the classification of anomaly detection in basic, statistical and probabilistic ways here, but there is way too much information around for me to figure out my next question.



      My question is:



      Which algorithm / minimal algorithms combination will [best] detect the kinds of abnormalities aforementioned and why?










      share|improve this question











      $endgroup$




      I aim to develop a generic algorithm to detect abnormalities in time series (ideally accepting multivariate series, and the presence of some noise). The aspects which I am considering for comparing these, are mainly:




      1. Seasonal variations

      2. Amplitude variations on periods.

      3. Period length variations.

      4. Delays (missing data) between periods

      5. Single points variations


      I have read about many algorithms, and the ones which I found more state of the art, are these:




      1. (year?) Using Deep learning autoencoders & decide period lengths via Fast Fourier Transform or ACF. ----- Article ----- Implementation

      2. (Apr 2017) Twitter STL + ESD algorithm. ----- Article ----- Implementation.

      3. (Nov 2017) Numenta HTM. Article. ---- Implementation


      Numenta also provides in this repository a benchmark between algorithms for anomaly detection. However, (1) I do not know if this benchmark englobes all main methods for this task, and (2) I am not sure if the way they did the benchmark include all the 5 abnormalities I mentioned.



      Also, I saw there are more algorithms, like the ones developed here, and I have seen the classification of anomaly detection in basic, statistical and probabilistic ways here, but there is way too much information around for me to figure out my next question.



      My question is:



      Which algorithm / minimal algorithms combination will [best] detect the kinds of abnormalities aforementioned and why?







      anomaly-detection






      share|improve this question















      share|improve this question













      share|improve this question




      share|improve this question








      edited 4 mins ago







      freesoul

















      asked Nov 12 '18 at 13:05









      freesoulfreesoul

      84




      84






















          0






          active

          oldest

          votes











          Your Answer





          StackExchange.ifUsing("editor", function () {
          return StackExchange.using("mathjaxEditing", function () {
          StackExchange.MarkdownEditor.creationCallbacks.add(function (editor, postfix) {
          StackExchange.mathjaxEditing.prepareWmdForMathJax(editor, postfix, [["$", "$"], ["\\(","\\)"]]);
          });
          });
          }, "mathjax-editing");

          StackExchange.ready(function() {
          var channelOptions = {
          tags: "".split(" "),
          id: "557"
          };
          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: false,
          noModals: true,
          showLowRepImageUploadWarning: true,
          reputationToPostImages: null,
          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


















          StackExchange.ready(
          function () {
          StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fdatascience.stackexchange.com%2fquestions%2f41101%2fwhich-between-a-dl-autoencoder-twitter-stlesd-and-numenta-hbm-is-more-state-of%23new-answer', 'question_page');
          }
          );

          Post as a guest















          Required, but never shown

























          0






          active

          oldest

          votes








          0






          active

          oldest

          votes









          active

          oldest

          votes






          active

          oldest

          votes
















          draft saved

          draft discarded




















































          Thanks for contributing an answer to Data Science Stack Exchange!


          • Please be sure to answer the question. Provide details and share your research!

          But avoid



          • Asking for help, clarification, or responding to other answers.

          • Making statements based on opinion; back them up with references or personal experience.


          Use MathJax to format equations. MathJax reference.


          To learn more, see our tips on writing great answers.




          draft saved


          draft discarded














          StackExchange.ready(
          function () {
          StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fdatascience.stackexchange.com%2fquestions%2f41101%2fwhich-between-a-dl-autoencoder-twitter-stlesd-and-numenta-hbm-is-more-state-of%23new-answer', 'question_page');
          }
          );

          Post as a guest















          Required, but never shown





















































          Required, but never shown














          Required, but never shown












          Required, but never shown







          Required, but never shown

































          Required, but never shown














          Required, but never shown












          Required, but never shown







          Required, but never shown







          Popular posts from this blog

          How to label and detect the document text images

          Vallis Paradisi

          Tabula Rosettana