What does 'Linear regularities among words' mean?












2












$begingroup$


Context: In the paper "Efficient Estimation of Word Representations in Vector Space" by T. Mikolov et al., the authors make use of the phrase: 'Linear regularities among words'.



What does that mean in the context of the paper, or in a general context related to NLP?



Quoting the paragraph from the paper:




Somewhat surprisingly, it was found that similarity of word
representations goes beyond simple syntactic regularities. Using a
word offset technique where simple algebraic operations are performed
on the word vectors, it was shown for example that vector(”King”) -
vector(”Man”) + vector(”Woman”) results in a vector that is closest to
the vector representation of the word Queen [20].



In this paper, we try to maximize accuracy of these vector operations
by developing new model architectures that preserve the linear
regularities among words. We design a new comprehensive test set for
measuring both syntactic and semantic regularities1 , and show that
many such regularities can be learned with high accuracy. Moreover, we
discuss how training time and accuracy depends on the dimensionality
of the word vectors and on the amount of the training data.







nlp language-model representation







share|improve this question









$endgroup$

















    2












    $begingroup$


    Context: In the paper "Efficient Estimation of Word Representations in Vector Space" by T. Mikolov et al., the authors make use of the phrase: 'Linear regularities among words'.



    What does that mean in the context of the paper, or in a general context related to NLP?



    Quoting the paragraph from the paper:




    Somewhat surprisingly, it was found that similarity of word
    representations goes beyond simple syntactic regularities. Using a
    word offset technique where simple algebraic operations are performed
    on the word vectors, it was shown for example that vector(”King”) -
    vector(”Man”) + vector(”Woman”) results in a vector that is closest to
    the vector representation of the word Queen [20].



    In this paper, we try to maximize accuracy of these vector operations
    by developing new model architectures that preserve the linear
    regularities among words    . We design a new comprehensive test set for
    measuring both syntactic and semantic regularities1 , and show that
    many such regularities can be learned with high accuracy. Moreover, we
    discuss how training time and accuracy depends on the dimensionality
    of the word vectors and on the amount of the training data.







    nlp language-model representation







    share|improve this question









    $endgroup$















      2












      2








      2





      $begingroup$


      Context: In the paper "Efficient Estimation of Word Representations in Vector Space" by T. Mikolov et al., the authors make use of the phrase: 'Linear regularities among words'.



      What does that mean in the context of the paper, or in a general context related to NLP?



      Quoting the paragraph from the paper:




      Somewhat surprisingly, it was found that similarity of word
      representations goes beyond simple syntactic regularities. Using a
      word offset technique where simple algebraic operations are performed
      on the word vectors, it was shown for example that vector(”King”) -
      vector(”Man”) + vector(”Woman”) results in a vector that is closest to
      the vector representation of the word Queen [20].



      In this paper, we try to maximize accuracy of these vector operations
      by developing new model architectures that preserve the linear
      regularities among words      . We design a new comprehensive test set for
      measuring both syntactic and semantic regularities1 , and show that
      many such regularities can be learned with high accuracy. Moreover, we
      discuss how training time and accuracy depends on the dimensionality
      of the word vectors and on the amount of the training data.







      nlp language-model representation







      share|improve this question









      $endgroup$




      Context: In the paper "Efficient Estimation of Word Representations in Vector Space" by T. Mikolov et al., the authors make use of the phrase: 'Linear regularities among words'.



      What does that mean in the context of the paper, or in a general context related to NLP?



      Quoting the paragraph from the paper:




      Somewhat surprisingly, it was found that similarity of word
      representations goes beyond simple syntactic regularities. Using a
      word offset technique where simple algebraic operations are performed
      on the word vectors, it was shown for example that vector(”King”) -
      vector(”Man”) + vector(”Woman”) results in a vector that is closest to
      the vector representation of the word Queen [20].



      In this paper, we try to maximize accuracy of these vector operations
      by developing new model architectures that preserve the linear
      regularities among words      . We design a new comprehensive test set for
      measuring both syntactic and semantic regularities1 , and show that
      many such regularities can be learned with high accuracy. Moreover, we
      discuss how training time and accuracy depends on the dimensionality
      of the word vectors and on the amount of the training data.







      nlp language-model representation




      nlp language-model representation






      share|improve this question













      share|improve this question











      share|improve this question




      share|improve this question










      asked 14 hours ago









      Dawny33♦Dawny33

      5,45183188




      5,45183188






















          1 Answer
          1






          active

          oldest

          votes


















          5












          $begingroup$

          By linear regularities among words, he meant that "Vectorized form of words should follow linear additive properties!"



          V("King") - V("Man") + V("Woman") ~ V("Queen)






          share|improve this answer









          $endgroup$













            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%2f46648%2fwhat-does-linear-regularities-among-words-mean%23new-answer', 'question_page');
            }
            );

            Post as a guest















            Required, but never shown

























            1 Answer
            1






            active

            oldest

            votes








            1 Answer
            1






            active

            oldest

            votes









            active

            oldest

            votes






            active

            oldest

            votes









            5












            $begingroup$

            By linear regularities among words, he meant that "Vectorized form of words should follow linear additive properties!"



            V("King") - V("Man") + V("Woman") ~ V("Queen)






            share|improve this answer









            $endgroup$


















              5












              $begingroup$

              By linear regularities among words, he meant that "Vectorized form of words should follow linear additive properties!"



              V("King") - V("Man") + V("Woman") ~ V("Queen)






              share|improve this answer









              $endgroup$
















                5












                5








                5





                $begingroup$

                By linear regularities among words, he meant that "Vectorized form of words should follow linear additive properties!"



                V("King") - V("Man") + V("Woman") ~ V("Queen)






                share|improve this answer









                $endgroup$



                By linear regularities among words, he meant that "Vectorized form of words should follow linear additive properties!"



                V("King") - V("Man") + V("Woman") ~ V("Queen)







                share|improve this answer












                share|improve this answer



                share|improve this answer










                answered 14 hours ago









                PreetPreet

                2563




                2563






























                    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%2f46648%2fwhat-does-linear-regularities-among-words-mean%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

                    GameSpot

                    Image
                    body.skin-minerva .mw-parser-output table.infobox caption{text-align:center} GameSpot 戰地風雲：惡名昭彰2在Gamespot的評論 网站类型 新聞 持有者 CBS 创始人 Pete Deemer Vince Broady Jon Epstein 网站 http://www.gamespot.com/ 注册 Optional (free and paid) 推出时间 1996年5月1日 [1] GameSpot （中国大陆：游戏基地），於1996年5月由Pete Deemer和Vince Broady創立，是一個提供新聞、評論、預告片、下載及其他的相關資訊的電子遊戲網站。GameSpot被一間後來被CNET收購的企業ZDNet所收購。根據Alexa，GameSpot.com是200個網路擁擠最嚴重的網站之一。 除了由GameSpot員工創作的內容，網站還允許用戶寫評論、網誌、之後在網路論壇分享。一些在CNET旗下的GameFAQs分享。 2004年， GameSpot被Spike TV的觀眾選上「電子遊戲賞節目」贏得「最傑出遊戲網站。 [2] 其他的遊戲網站還有IGN、1UP.com、GameSpy是它最大的競爭對手。2008年，根據Compete.com的統計，「gamespot.com」吸引了最少6000萬人的點擊率。 [3] GameSpot的主頁鏈結了到最近新聞、評論、預告、和一些有關遊戲機的入口：Wii、任天堂DS、電腦遊戲、Xbox 360、PSP、PlayStation 2、PlayStation 3。它還有一列「最受歡迎遊戲名單」，還有給用戶快速獲得遊戲資訊的搜尋器。GameSpot 還包括一些小範圍的遊戲機：任天堂64、GameCube、Game Boy Color、Game Boy Advance、Xbox、PlayStation、SEGA Saturn、Dreamcast、Neo Geo Pocket Color、N-Gage、手機遊戲。 目录 1 歷史 1.1 國際歷史 1.2 著名的員工 2 評論和分...
                    Read more

                    日野市

                    Image
                    body.skin-minerva .mw-parser-output table.infobox caption{text-align:center} 日野市 日野市 日文轉寫  • 日文 日野市  • 平假名 ひのし  • 罗马字 Hino-shi 金剛寺（關東三十六不動尊靈場第9號）的不動堂 市旗 徽章 日野市在東京都的位置 日野市 日野市在日本的位置 坐标： 35°40′17″N 139°23′42″E  /  35.6714°N 139.395°E  / 35.6714; 139.395 国家   日本 地方 關東地方 都道府縣 東京都 接鄰行政區 府中市、國立市、立川市、昭島市、八王子市、多摩市 政府  • 市長 大坪冬彦 面积  •  总计 27.53 平方公里（10.63 平方英里） 人口 （2014年12月1日）  • 總計 183,323  • 密度 6,660/平方公里（17,200/平方英里） 象徵  • 市樹 樫  • 市花 菊花  • 市鳥 普通翠鳥 时区 日本標準時間 （UTC+9） 地方公共團體編號 13212-8 邮政编码 〒 191-8686 市役所地址 日野市神明一丁目12番1號 電話號碼 +81-42-585-1111 法人編號 1000020132128 網站 http://www.city.hino.lg.jp/ 人口：日野市官方網頁 日野市 （日语： 日野市 ／ ひのし   Hino shi   * / ? ）為一位于東京都（不含島嶼部分）中央地帶的城市。從東京站乘坐中央線特別快速列車45分鐘即可到該市。面積27.53km²。1963年（昭和38年）11月3日，該市開始實行市制，為全日本第559個市。汽車企業日野自動車的總部位於這裡。 往東京都特別區部的通勤率為20.9%，往八王子市的通勤率為12.0%（平成22年國勢調査）。 目录 1 概要 ...
                    Read more

                    Tu-95轟炸機

                    Image
                    body.skin-minerva .mw-parser-output table.infobox caption{text-align:center} 图-95 Ту–95 北约代号：熊（Bear） 一架在安格斯空军基地的图-95MS 概觀 類型 战略轰炸机 代號 北约代号： Bear （熊） 乘員 7名 駕駛員2名，機尾炮手1名，其他人員4名 首飛 1952年11月12日 服役 1956年 設計 圖波列夫設計局 產量 超过500架 現況 现役 主要用戶   蘇聯   俄羅斯 衍生機型 图-114、图-119、Tu-95 技术数据 長度 49.50米（162呎5吋） 翼展 51.10米（167呎8吋） 高度 12.12米（39呎9吋） 翼面積 310平方米（3,330平方呎） 空重 90,000公斤（198,000磅） 最大起飛重量 188,000公斤（414,500磅） 發動機 4具库兹涅佐夫NK-12MV型渦輪螺旋槳發動機 功率 4×11,000千瓦（14,800馬力） 性能數據 最大速度 925公里／時（500節，575哩／時） 爬升率 10米／秒（2,000呎／分） 最大升限 12,000米（39,000呎） 最大航程 15,000公里（8,100海浬，9,400哩） 翼負荷 606公斤／平方米（124磅／平方呎） 推重比 235 W／公斤（0.143馬力／磅） 武器装备 機炮 1具/2具AM-23型23公釐雷達控制機砲（機尾） 飛彈 空对地导弹： Kh-20、Kh-22 Kh-26、Kh-55 炸彈 60枚FAB-250炸彈，或 30枚FAB-500炸彈。 其他 载弹量（包括飞弹）：最多15,000公斤（33,000磅） Tu-95熊式MR型機 图-95 （俄语： Ту–95 ， 英语： Tu-95 ），北約代號： Bear （ 熊 ），是蘇聯圖波列夫設計局所研製，是全世界唯一服役的大型四渦輪螺旋槳發動機...
                    Read more