There exist already many music search engines. However, they are all asking the name of a song or a singer.
The problem is that many people are searching for the songs that they have just heard and they don't know exactly the name of the group or the song. So there is a demand for a special music search engine based on new principles.
Interesting approach to a special-purpose music discovery search engine was proposed by the "Herd It" team: search by description.
The search engine is only in Beta version, however there is a possibility to find a music by style, mood (happy, sad, energetic), by color with what you associate this music.
For creating such kind of music tagging that enables this new concept of music search human computation is required.
So in traditions of "games with a purpose" a multi-player online game, positioned on Facebook, was developed to enable the crowdsourcing.
The game is easy and fun to play. I find it the most interesting games of all described. The design is also quite awesome!!!
Am really excited about the future release of this beta-search.
Saturday, February 6, 2010
Tuesday, December 15, 2009
Some important publications about "games with a purpose"
- L. von Ahn: “Games with a purpose”. Computer, vol. 29(6), pp. 92–94, 2006.
- L. von Ahn and L. Dabbish: “Labeling images with a computer game”. In proceedings of conference human factors in computing systems (CHI 04), ACM Press, pp. 319–326, 2004.
- L. von Ahn: “Peekaboom: A game for locating objects in images”. In proceedings of conference human factors in computing systems (CHI06), ACM Press, pp. 55–64, 2006.
- E. Law et al.: “Tagatune”. In proceedings of International Conference Music Information Retrieval (ISMIR 07), Austrian Computer Soc., pp.361–364, 2007.
Monday, November 23, 2009
Sentiment Quiz: Playing On Facebook
A good way to find people for playing your game is to put it on the social network: And what can be better and more effective than Facebook?!
Sentiment Quiz exploits human intelligence for providing words' collocations. A player should choose if the word written is positive, very positive, negative, very negative or neutral. If his choice matches with the answers of other players, he will get points and a small icon depicting him tries to get ahead of his concurents. The game is available in 7 languages: English, German, French, Spanish, Italian, Portuguese and Russian. One can win DVD from movie collection if he gets the most points.
Design of the game leaves somehow much to be desired. In comparison with the GWAP, it is not so colourful and competetive. There is also no time restriction: you can play as long as you wish.
The results will be used for some scientific research. I think, something in SentiWordNet style or opinion mining. Maybe also lexical functions or machine translations.
Somehow, the Sentiment Quiz team have recently gathered 98% evaluated terms. Within this period 5540 players helped evaluating terms from different languages!!! Sounds amazing!!!
Sentiment Quiz exploits human intelligence for providing words' collocations. A player should choose if the word written is positive, very positive, negative, very negative or neutral. If his choice matches with the answers of other players, he will get points and a small icon depicting him tries to get ahead of his concurents. The game is available in 7 languages: English, German, French, Spanish, Italian, Portuguese and Russian. One can win DVD from movie collection if he gets the most points.
Design of the game leaves somehow much to be desired. In comparison with the GWAP, it is not so colourful and competetive. There is also no time restriction: you can play as long as you wish.
The results will be used for some scientific research. I think, something in SentiWordNet style or opinion mining. Maybe also lexical functions or machine translations.
Somehow, the Sentiment Quiz team have recently gathered 98% evaluated terms. Within this period 5540 players helped evaluating terms from different languages!!! Sounds amazing!!!
Wednesday, November 18, 2009
New image search by Google: Google Image Swirl
Google Image Swirl Beta is a new Google's approach for the image search.
Images are grouped in cycles, or "swirls" according to the similarity and concept relations. Google pretends to "go beyond just relying on text", as a Google Image Swirl Product Manager Aparna Chennapragada announced in the interview with the eWeek. Now Google Image Swirl search only for 200.000 queries but this amount is said to be increasing with the time. Google exploits object recognition algorithms together with the extracting of textual information about the image.
I like especially the design. It would be great to use Google' s search with Touchscreen.
Bing is said in both articles about Google Image Swirl (eWeek and ReadWriteWeb) to be a concurrent and actually a leader in the image search. But I found Bing VisualSearch quite different and thus incomparable with the Google concept.
You can try Bing VisualSearch only from Internet Explorer, only if you install Silverlight and only in English. More about Bing Visual Search you can find here:
http://www.readwriteweb.com/archives/bing_launches_visual_search.php
However according to statistics of using the browsers: Mozilla is currently leading. In the July 31st Mozilla passed 1.000.000.000 downloads.
Images are grouped in cycles, or "swirls" according to the similarity and concept relations. Google pretends to "go beyond just relying on text", as a Google Image Swirl Product Manager Aparna Chennapragada announced in the interview with the eWeek. Now Google Image Swirl search only for 200.000 queries but this amount is said to be increasing with the time. Google exploits object recognition algorithms together with the extracting of textual information about the image.
I like especially the design. It would be great to use Google' s search with Touchscreen.
Bing is said in both articles about Google Image Swirl (eWeek and ReadWriteWeb) to be a concurrent and actually a leader in the image search. But I found Bing VisualSearch quite different and thus incomparable with the Google concept.
You can try Bing VisualSearch only from Internet Explorer, only if you install Silverlight and only in English. More about Bing Visual Search you can find here:
http://www.readwriteweb.com/archives/bing_launches_visual_search.php
However according to statistics of using the browsers: Mozilla is currently leading. In the July 31st Mozilla passed 1.000.000.000 downloads.
Sunday, November 15, 2009
GWAP from Bing: Page Hunt
Page Hunt is investigating the human search behavior. Launched in July 2009, the online game presents users ('page hunters') with a random Web page asking them to input search terms that will place this Website within Bing’s top five search results.
If the search query puts the page on the first place, then the user gets 100 points, on the second place - 90 points. For non-frequent successful queries, the player gets additional scores.
I find disadvantageous that there is no explicit competition as the user plays alone. I suppose, if the online game would be multiplayer, it would bring additional motivation for the user.
More information about the game can be found:
on eWeek;
on ReadWriteWeb;
and on Technology review.
A publication to this topic:
H. Ma, R. Chandrasekar, C. Quirk, A. Gupta: “Page Hunt: Improving search engines using human computation games”. Proceedings of the 32nd international ACM SIGIR conference on Research and development in information retrieval, 2009.
If the search query puts the page on the first place, then the user gets 100 points, on the second place - 90 points. For non-frequent successful queries, the player gets additional scores.
I find disadvantageous that there is no explicit competition as the user plays alone. I suppose, if the online game would be multiplayer, it would bring additional motivation for the user.
More information about the game can be found:
on eWeek;
on ReadWriteWeb;
and on Technology review.
A publication to this topic:
H. Ma, R. Chandrasekar, C. Quirk, A. Gupta: “Page Hunt: Improving search engines using human computation games”. Proceedings of the 32nd international ACM SIGIR conference on Research and development in information retrieval, 2009.
Labels:
Bing,
games with a purpose,
gwap,
IR,
online games,
search
The Web knows what you feel
"We feel fine!" explors human emotions on the Web. If you have written in your LiveJournal, MSN Spaces, MySpace, Blogger, Flickr, Technorati, Feedster, Ice Rocket, etc. what you feel or what you are feeling in this moment, your feeling could also be crawled by "We feel fine!"
15000-20000 new feelings a day and over 12 million feelings since 2005.
"We feel fine" contains about 5000 pre-defined feelings. When the crawler finds a phrase "I feel/ I am feeling", it records the full sentence, up to the period, and identifies the "feeling" expressed in that sentence (e.g. sad, happy, depressed, etc.). Because blogs are structured in largely standard ways, the age, gender, and geographical location of the author can often be extracted and saved along with the sentence, as can the local weather conditions at the time the sentence was written as well as images. All of this information is saved and then the statistics about feelings is provided.
Now you can even buy the book "We Feel Fine: An Almanac of Human Emotion"
This is an approach that was inspired by techniques used in Listening Post, a wonderful project by Ben Rubin and Mark Hansen.
Another collection of tools investigating the mood on the Web is provided by MoodViews
15000-20000 new feelings a day and over 12 million feelings since 2005.
"We feel fine" contains about 5000 pre-defined feelings. When the crawler finds a phrase "I feel/ I am feeling", it records the full sentence, up to the period, and identifies the "feeling" expressed in that sentence (e.g. sad, happy, depressed, etc.). Because blogs are structured in largely standard ways, the age, gender, and geographical location of the author can often be extracted and saved along with the sentence, as can the local weather conditions at the time the sentence was written as well as images. All of this information is saved and then the statistics about feelings is provided.
Now you can even buy the book "We Feel Fine: An Almanac of Human Emotion"
This is an approach that was inspired by techniques used in Listening Post, a wonderful project by Ben Rubin and Mark Hansen.
Another collection of tools investigating the mood on the Web is provided by MoodViews
Thursday, November 12, 2009
Who is donating to Wikipedia?
Donations to Wikipedia have exceeded 6 Mio in Januar this year after the pleaing of the Wikipedia's founder Jimmy Wales. Wikipedia was launched in 2001, so it is quite a good sum for the non-profit project, I suppose.
The motto on the Wikimedia page sounds somehow a bit pathetic:
"It stopped being just a website a long time ago. For many of us, most of us, Wikipedia has become an indispensable part of our daily lives."
Help protect it now. Please make a donation today.
It reminds strongly GreenPeace ads.
I am wondering who is actually donating money. I mean, private persons. Why donating to Wikipedia? Why not better help children in Africa? Are they impressed by the slogan? By the project?
I was trying to understand the idea. Here you can have a look at the list.
Donators leave their name quite often with the city names, so i have found some of them afterwards by Google. I am wondering about the persons who spend more than 50$ to Wikipedia. For example, there are some private persons who are spending about 100$, like Matthias Graf from Magdeburg or Christian Bahls from Rostock. Some people want to make SEO, they leave the links to their websites. but most of them are just saying "thanks" to Wikipedia.
People want to thank Wikipedia for such brilliant project, for much awesome information!
but I mean the information written on Wikipedia is written by other users for free, so you can say "thank you" by writing another article or translating. or you can donate the money for the person who has written the article that helped you. Then it would be more fair. Maybe this Matthiases and Christians don't understand that fact?
And how do you think?
The motto on the Wikimedia page sounds somehow a bit pathetic:
"It stopped being just a website a long time ago. For many of us, most of us, Wikipedia has become an indispensable part of our daily lives."
Help protect it now. Please make a donation today.
It reminds strongly GreenPeace ads.
I am wondering who is actually donating money. I mean, private persons. Why donating to Wikipedia? Why not better help children in Africa? Are they impressed by the slogan? By the project?
I was trying to understand the idea. Here you can have a look at the list.
Donators leave their name quite often with the city names, so i have found some of them afterwards by Google. I am wondering about the persons who spend more than 50$ to Wikipedia. For example, there are some private persons who are spending about 100$, like Matthias Graf from Magdeburg or Christian Bahls from Rostock. Some people want to make SEO, they leave the links to their websites. but most of them are just saying "thanks" to Wikipedia.
People want to thank Wikipedia for such brilliant project, for much awesome information!
but I mean the information written on Wikipedia is written by other users for free, so you can say "thank you" by writing another article or translating. or you can donate the money for the person who has written the article that helped you. Then it would be more fair. Maybe this Matthiases and Christians don't understand that fact?
And how do you think?
Subscribe to:
Posts (Atom)