اگر اين دو بازيکن، کلماتی يکسان را به تصويری يکسان اختصاص دهند، امتياز می‌گيرند. اما در پشت پرده چيزی فراتر از يک بازی سرگرم‌کننده پنهان شده است.

شما در واقع دارید به بهينه‌سازی جست‌وجوی گوگل کمک می‌کنيد. شما بازی می‌کنيد اما گوگل در حين بازی از قوه تشخيص شما به نفع جست‌وجويی بهتر بهره‌برداری می‌کند و کليدواژه هايش برای تصاوير مختلف را گسترش می دهد.

گوگل که لقب «غول جست‌وجوی اينترنت» را يدک می‌کشد، سعی می‌کند با انواع و اقسام روش‌ها نتايجی دقيق‌تر برای جست‌وجوها عرضه کند. يکی از دلايل خريد سايت‌هايی چون «يوتيوب» و «بلاگر» توسط گوگل و «خوشمزه» و «فليکر» توسط ياهو برای فهرست کردن هر چه سريع‌تر محتوای روی اينترنت است تا نتايج جستجو را با دقت بيشتری به کاربران ارائه دهند.

اما نقطه ضعف همه موتورهای جست‌وجو در جست‌وجوی تصاوير است. موتورهای جست‌وجو در ارائه نتايج دقيق‌تر در جست‌وجوی تصاوير با هم رقابت تنگاتنگی دارند.

اساساً روبات‌های جست‌وجوگر اين موتورها مشکل چندانی با يافتن و فهرست کردن متون ندارند اما برای فهرست کردن تصاوير تنها به «برچسب‌»ها و تکنيک‌های «هوش مصنوعی» در تشخيص تصاوير متکی هستند.

اين وصله‌ها به کی می‌چسبد؟

اصطلاحی که اين روزها با نام «برچسب» يا label باب شده در حقيقت يک همکاری دسته جمعی، برای رسيدن به آينده‌ای بهتر در جست‌وجوی اينترنتی است.

با ايجاد مفهومی به نام وب ۲، برچسب‌ها نيز پای خود را به وبلاگ‌ها، فتوبلاگ‌ها و سايت‌های مديريت محتوا باز کردند. اين روزها توليدکنندگان محتوا با قرار دادن کلماتی کليدی به عنوان برچسب در پای هر نوشته يا عکس به تدريج نوعی دسته‌بندی موضوعی ايجاد می‌کنند که در حقيقت به نفع کاربران است.

کاربر چنين سايتی می‌تواند با مشاهده يک متن يا عکس، متن‌ها و عکس‌های مرتبط با آن را به سادگی بيابد. موتورهای جست‌وجو هم با تغييراتی در روش خود به اين برچسب‌ها حساس شده‌اند و روبات‌های جست‌وجوگر ارتباطی منطقی با اين برچسب‌ها پيدا کرده‌اند. نکته اینجاست که اين وصله‌ها به هر کسی اگر نچسبد، به گوگل حسابی می‌چسبد.

هوش از نوع ماشينی

تکنيک OCR در هوش مصنوعی در واقع استفاده از روش‌هايی است که شامل پردازش محتوای تصاوير می‌شود.

بگذاريد برايتان مثالی بزنم. شما دو عکس ديجيتالی داريد که يکی از آنها منظره‌ای از جنگل است و ديگری صفحه‌ای اسکن شده از يک کتاب. برای کامپيوتر به عنوان يک ماشين هر دوی آنها تصاويری هستند متشکل از نقاط رنگی (پيکسل) و هيچ مفهومی ندارند.

در حقيقت اين شما هستيد که تشخيص می‌دهید کدام يک منظره است و کدام يک متن، و کامپيوتر از اين تشخيص عاجز است.

شاخه‌ای از هوش مصنوعی (به عنوان يکی از علوم انفورماتيک) OCR ناميده می‌شود که تلاشی است برای پردازش تصاوير تا کامپیوتر را واجد چنين تشخيصی کند. اما در بسياری از موارد ماشين باز هم در اين شناخت ناموفق است. اين جا است که کاربران به عنوان کارگران تعليم‌دهنده وارد عمل می‌شوند.

تلاش‌های اخير بر تعليم کامپیوترها استوار است و اين چيزی نيست به جز واداشتن کامپيوترها به فکرکردن مانند انسان. اين شيوه را «تشديد هوش» يا Intelligence Augmentation نام داده اند. اما پشت اين عنوان پرهيبت چيزی نيست جز يک ترفند زيرکانه دلنشين يعنی: کلاه گذاشتن!

در اين روش انسان به حل بخش‌هايی خاص از يک مسأله وادار می‌شود. حل چنين مسأله‌ای برای کامپيوتر مشکل يا ناممکن اما برای انسان پيش پا افتاده است. همزمان، با اين کار کامپيوتر تعليم می‌بيند تا بيشتر شبيه به انسان عمل کند.

خلاصه آن که شما با «برچسب‌زنی گوگل» در حال بازی هستيد اما کلمه‌ای را که به عنوان برچسب به يک عکس اختصاص می‌دهيد و با برچسب هم‌بازی شما منطبق می‌شود به عکس مورد نظر وابسته می‌شود. از آن ‌به بعد آن تصوير در نتايجی ظاهر خواهد شد که با آن کلمه جست‌وجو شده‌اند.

شما در ظاهر بازی می‌کنيد اما در باطن روبات‌های گوگل را آموزش می‌دهيد که اين عکس «پرنده»ای «دريا»يی است در حال «پرواز».

روش تشديد هوش، دايره گسترده‌ای را در بر می‌گيرد و گوگل تنها بازيگر اين عرصه نيست. «کپچاهای ترجمه» نمونه‌ای جالب‌تر از اين روش است که درباره‌اش خواهيم نوشت.

منبع : نیما اکبرپور - زیگ زاگ 

About Google Image Labeler

Google Image Labeler is a feature, in form of a game, of Google Image Search that allows the user to label random images to help improve the quality of Google's image search results. 

History 

On August 31, 2006 Google launched this service, as a beta.

Luis von Ahn developed the ESP Game, a game in which two people are simultaneously given an image of the same picture, with no way to communicate. The ESP Game has been licensed by Google in the form of the Google Image Labeler.

Players have noticed various subtle changes in the game over time. In the earliest months, through about November, 2006, you could see your partner's guesses during play by mousing over the image. When the "congenita abuse" started (see below) you could see if your partner was using those terms, while the game was underway. The game was changed so that only at the end of the game could you click "see partner's guesses" and learn what he had typed. "Congenita abuse" was finally stopped by changes in the structure of the game in Feb. 2007 (see below). During the first few months of 2007 regular players grew to recognize a group of images that signified a "robot" partner, always with the same labels in the same order. This appears to have changed as of about March 13, 2007. Suddenly most of the images seen are brand new, and the older images come with extensive off-limits lists.

As of May 4, 2007 there have been fundamental and substantial changes made to the game. Instead of 90 seconds, players now have 2 minutes. Instead of 100 points per image, the score is varied to reward higher specificity. "Man" might get 50 points whereas "Bill Gates" might get 140 points.

 Rules

The user will be randomly paired with a partner who is online and using the feature. Users can be registered players who accumulate a total score over all games played, or guests who just play for one game. Note that players come from around the world, some practicing their English, and both American and British English will be encountered (soccer vs. football). When an uneven number of players are online, a player will play against a prerecorded set of words.

As of May 4, 2007, some rules have changed. Over a 2 minute period, (formerly 90 seconds) the user and his/her partner will be shown the same set of images and asked to provide as many labels as possible to describe each image you see. When the user's label matches the partner's label, both will earn points and move on to the next image until time runs out. It is possible to pass on an image but both users must agree to do this. Matched images used to earn 100 points for both partners. Now the score is variable from 50 to 150 depending on the specificity of the answer. The 150 score is rare but 140 points will be awarded for a name or specific word, even if the word is spelled out in the image. Terms with low specificity like "trees" or "man" earn only 50 points. There has never been any screening for correctness, so that if both players type "Jupiter" for an image of Saturn, they would presumably both get 140 points.

Labels that have been agreed on by previous users may show on an "off limits" list and cannot be used in that round. Some players think that the game staggers appearance of the images, and that sometimes it takes the first words typed by one player to form an "off limits" list for the other player. In other words, the off limits words may be unilateral, asymmetrical. This would explain the rather frequent circumstance when it seems a partner can't think of words like "car," "bird," or "girl." Very rarely, at the end of the match it becomes obvious that one image was different for the two players. Perhaps this is simply an error, or perhaps it is a test to see how quickly people will pass when their descriptions do not match. At times, one user's computer will fail to load a new image, or continue to display the previous image shown. Times likes these also call for a mutual "pass" on the part of both players.

After the 120 second time expires, the user can see the user name of the partner, their score (with which both are credited), their cumulative score to date and a view of the images they have matched or agreed to pass in that round. In the new version of the game, the last image (not matched) is also retained but marked as "passed."

The images themselves then link to the websites where those images were found and can be explored for more information and to satisfy curiosity.

All the partner's label attempts are revealed at the end of the game, which can help to understand what their perceptions of the images were.

Top Scorers 

The game keeps the high scores of registered users and these are displayed both for the day and for "all time". The highest all time scorer for the first months was wordgirl -- the first player to pass 1,000,000 points. She stopped playing in early November 2006, and was passed by "no random words" by the end of the month. "No random words" passed 2,000,000 points and led until FairlandOklahoma appeared with an enormous score. "FairlandOK" reached 10,000,000 points, and then changed her name to "Night Elf Rogue."

Google is betting on users' competitiveness to rack up high scores to swell the number of images ranked.

Benefits to Google 

The game is not designed simply for fun. Though the feature is enjoyable for the users, it is also a clever way for Google to ensure that its keywords are matched to correct images. Each matched word will help Google to build an accurate database used when using the Google Image Search.

Without human tagging of images, Google Images search has in the past relied on the context of the image. For example, a photo that is captioned "Portrait of Bill Gates" might have "Bill Gates" associated as a possible search term. The Google Image Labeler relies on humans that tag the meaning or content of the image, rather than its context looking on at where the image was used. By storing more information about the image, Google stores more possible avenues of discovering the image in response to a user's search.

Issues and Problems 

Additional Rules

    * Some users complain that the rules are difficult to decipher. Nowhere is it stated, for example, that a player should press Return after typing a label.
    * Beginners often make the mistake of typing several terms into the first box, not realizing that those words are then all considered, together, as a phrase.
    * The "pass" option is also not explained; although it means that the player does not want to rate a word, some players have thought that this button is to be pressed after making a guess. Either of these mistakes can easily result in a zero score.

While these rules are not explained on the playing screen, some are in fact explained on the Help Page.Other issues include:

    * Some images may fail to load, or will load very slowly, using time off of the clock
    * Experienced users type the letter "x" to avoid simply passing (and scoring zero) when this happens. Note that "x" (nor any other word) will not work for two successive images so other terms ("blank" or "none") could also be utilized, since sometimes many images do not load.
    * Only six labels appear on the screen during a round, if more are added some scroll off the top. Observation shows that all of these labels count, and with a fast partner you can see nine or ten words for one image. Also, after hitting "Pass" you will no longer see the count of your partner's labels. If you continue to type, you can get a match on words your partner had already typed when you passed. If your partner types a matching word after you pass, it will not count.
    * Scores are typically low when several of the images presented have a number of "off limits" tags. In some cases, the "off limits" tags are quite extensive or exhaustive, making it difficult for both partners to create a novel tag that matches. This tends to use time and reduce the total score.

Abuse 

Less than a month after the launch, the game began to be abused . It appeared that Google was getting spammed with words from the following list: abrasives, accretion, bequeathing, carcinoma, congenita, diphosphonate, entrepreneurialism, forbearance and googley. Because players can see the responses of their partner at the end of each round, they learned that other players were using these words. Some then incorporated these words into their answers for entertainment, a sense of social belonging, or simply to increase their score. As of Feb. 7, 2007 Google changed the game to cut down on the abuse. The words on the list above were filtered out. Also certain images that had become triggers for the random words came with an immediate "Your Partner Wants to Pass." In the game revision of March 13, 2007, the "trigger images" were removed for a while but they are back in play as of March 26.