Can Twitter be used as viral marketing?

I’ve been redesigning my website – www.sakcreations.co.uk for the last couple of weeks, it was long overdue an update.   Sak Creations has been my Web Design moniker for the last 10 years, but as web design has now become a sideline to my PhD I’ve also put a section up to be the formal home of my research interests.   Leaving this blog to be the home of my infrequent ramblings.

I’ve been giving some thought to SEO whilst redesigning my website,  also sparked by a discussion with my graphic designer friend KJ Creative about how to improve google rankings.  The main reason I’ve kept my research on my Sak Creations domain (as well as the fact I have hosting and everything set up already and I don’t really want to have my identity pinned on Stacey Greenaway, in case I don’t always stay Stacey Greenaway) is that it is creeping up the page rank scale (slowly) and Google know to return Sak Creations if someone types my name in.   So shameless backlinks aside, I have given thought to how you promote websites nowadays when you can’t compete with just good meta information, textual content and the good old reciprocal link.  It seems any designer creeping up the page ranks does so by promoting themselves and their work on various forums and designer community websites, commenting on blogs, writing blogs, twittering about their work.   This social web gives a whole new depth to viral marketing.  I remember sitting in meetings discussing how to market our current projects in viral emails, is it as simple now as just twittering about it and posting a link on facebook?  Well maybe if you have enough followers and friends.   So here’s the next question, how do you get all the followers and friends in order to make you viral marketing/shameless self promotion a success?  Do you have to spend more time twittering about your work than actually doing the work?  Does a marketing assisitants job spec now include running the companies twitter account?

There are more problems with relying on social web to improve those rankings, first to engage in the social web and make it work for you, you have to be a social person and second you need to invest the time on the sites ‘networking’ instead of doing the work.  Is social media really only useful for people with ALOT of time on their hands, or for big companies who can afford to employ someone to be the ‘social’ face of the company/organisation?

So, as a person with not enough time on my hands, but a desire to find new ways to avoid doing the actual work of my PhD by finding tasks that can loosely be deemed as work;  I have decided to succumb to the power of Twitter and set up an account for VideoTag.  I can start twittering about the game and ideas and struggles during development, then hopefully have some followers who want to play the game when it launches.  then I can continue to use it as a vehicle to launch new games.

I’d like to just add another problem to using social media to promote yourself/project – usernames, they are always taken by someone!!! Mr jesse videotag isn’t even using his account!!!  So you can follow videotag2 instead.  Now comes the hard work knowing what to twitter about!

**Is part of Twitters appeal the cute little bird?

VideoTag experiment – Results

The big plan had been to write up my MSc project – VideoTag into a paper and get it published.  It has been over 12 months now since I finished this experiment and my ideas have moved on,  I am all to aware of the short comings of the project in terms of academic paper.  And I guess, if I am going to publish something, I want to be proud of it.  Not that I’m not proud of VideoTag, but it needs a lot of improving.  I have so many ideas of where I can push this concept for my PhD, that I want to spend my time doing that, rather than re-writing something I did 12 months ago.

However, I figured it would be a shame to never have the results of the experiment somewhere, as I did find out some stuff!!  I found out enough to warrant me being able to continue researching the concept of video tagging games for the next 3 years.

So here they are, mostly for my own benefit so I have somewhere to reference the experiment if necessary in future work.  I am sure my traffic will go through the roof with interest in these!!

VideoTag – A game to encourage tagging of videos to improve accessibility and search.

Results

Data Collection/Tag Analysis
Data Collection
Usage was monitored over a month long period, after which the dataset for analysis was downloaded. Data was analysed for the period July 30th 2007 – September 2nd 2007. Data was also included from the user testing phase, July 15th 2007 – July 30th 2007

Tag Analysis
Quantitative methods of evaluating the VideoTag data involved analysis of sets of tags for a Zipf distribution on a graph.  Furnas et al. (1998) discuss how power law distribution demonstrates the 80-20 rule.  While 20% of the tags have high frequency and therefore a higher probability of agreement on terms, 80% have low frequency and corresponding low probability.  When the tags are inspected for cognitive level, Cattutto et al. (2007) discovered that high rank, high frequency tags are of basic cognitive level, where as low rank, low frequency tags are of subordinate cognitive level.  In terms of analysing VideoTag data this is a useful method to determine whether the game has been effective at improving the specificity of tags, in order that a greater number of subordinate level tags can create user descriptions of video to improve video accessibility.  Following the 80-20 rule, an overall increase in the amount of tags would increase the amount of basic level tags that have a high probability of agreement on terms, improving video search.  Drawing comparisons between the quantity of tags per video generated through VideoTag in relation to tags per video assigned in YouTube, will indicate whether VideoTag has been successful at increasing the amount of tags and therefore could be a useful tool at improving video search.  A paired t-test analysis was conducted to statistically qualify the comparison results.

Tag type was evaluated using qualitative methods, both YouTube and VideoTag tags were analysed for evidence of Golder and Huberman (2005) tag types and comparisons drawn.

Results
General usage analysis revealed that 243 games were played in total during the experimental period. 87 of those games were discounted because the game points score was 0, indicating that the players had not tagged any videos. 37 of those discounted games were guest users (i.e. who did not log in). The 156 valid games were played by 96 unique users, meaning that some users played more than one game. Of the 96 unique users, 73 were registered, 23 were guests.

Table 1 % of tags entered ordered by Tagging Support

Table 1 % of tags entered ordered by Tagging Support

Blind tags, as defined by Marlow et al. (2006), are free tags entered without prompting the user with suggestions (suggested tagging). Guided tagging, as introduced by Bar Ilan et al. (2006) gives structure to tagging by offering the user guidelines. During the 156 valid games, a total of 4490 tags were entered. 4076 of these were Blind, 68 were Guided and 346 were Pitfalls (Fig 1). The substantial preference for blind over guided tagging means that the tag data generated by VideoTag in this experiment can not be used to compare the cognitive levels of blind or guided tags.  Tag analysis therefore compared blind tags and pitfalls and omitted the suggestion differential.

Fig 2 Frequency of blind tags per video

Fig 1 Frequency of blind tags per video

The long tail effect apparent when blind tag frequencies are plotted on a graph, fig 1, has evidence of a Zipf distribution.  The vast majority of tags, 52.9 % occur only once. In relation to the research findings of Cattuto et al. (2007) and Golder and Huberman (2005), fig 1 would suggest that VideoTag generates an increased number of subordinate level, descriptive tags over basic level tags of high frequency.

Fig 2 Frequency of pitfalls per video

This finding is emphasised when examining fig 2 which plots the pitfall frequency. Pitfalls were created as basic cognitive level; they were the tags that were expected to have the least cognitive cost. It was expected that these tags would have high frequency, as they would be the tags that came to a players mind first. Few low frequency tags were expected if the basic level tags for each video had been predicted successfully. Partial success is shown, with only 20.23% of pitfalls occurring once. This occurrence can be explained by the fact that the majority of users played level one. Inspection of the tag data (Table 2 provides an example of tag data for a video in level 1) revealed that the majority of high frequency pitfall tags were assigned to videos in Level 1, the most played level. Therefore it could be expected that if the levels of VideoTag had been played more evenly then there would be a lower percentage of low frequency tags.

The low amount of pitfall tags (346 out of 4490) coupled with the high frequency of low frequency blind tags, is an indication of the success of the gameplay element of encouraging users to avoid pitfalls and enter more subordinate cognitive level tags. It has contributed to VideoTag’s effectiveness as a tool for generating more descriptive tags.

Fig 3 Frequency of all tags entered during the VideoTag experiment, grouped by video

This is further implied by analysing the frequency of all tags entered, as shown in fig 3. This graph shows a clear long tail effect, with the majority of tags entered having low frequency. Whilst the high frequency tags are useful for video search, because agreement on terms will be reached quicker, the low frequency tags are important as there is a likelihood that out of the billions of internet users, at least one other user will agree on a term. The amount of tags entered, and the high amount of low frequency tags implies that VideoTag has been successful at generating a large amount of high quality tags that can be used to create descriptions of the videos for visually impaired users. If high agreement was present for all tags, then there would not be enough variety in the tags to sufficiently create the descriptions.

This result is further evident in Fig 4, which represents the frequency of all tags frequency and shows the appearance of a power law (i.e. a straight line in a log-log scale). There is a greater frequency of low frequency tags, which is a pleasing result as this was the main aim of the project, to encourage users to tag videos with more descriptive tags. By generating more low frequency, subordinate level tags, more useful descriptions can be created to improve internet video accessibility.

Fig 4 Frequency of tag frequencies for all tags entered during the experiment.

Analysis suggests that VideoTag has been successful at increasing the amount of tags entered for each video. Geisler, G. and Burns, S. (2007) found the average amount of tags on YouTube to be 6. The average number of tags per video in VideoTag compares very favourably at 71.3. Fig 5 clearly shows this increase in tags generated by VideoTag compared to the tags entered for each video in YouTube. The tags are grouped by Video Id and ordered by levels 1-5 ascending. An interesting anomaly in the graph shows an increased number of tags for one video at each level, indicating that out of a purely random selection, one video was selected more times. With the graph alone it can be said that VideoTag created more tags for videos than are entered on YouTube, which could be beneficial to both video search and accessibility. A paired t-test of the amount of tags entered both on VideoTag and YouTube returned a p value of 0.000 which shows that this difference is statistically significant, proving that a game environment encourages more tags for videos.

Fig 5 Amount of YouTube tags per video compared to the amount of VideoTag tags, ordered by Level.

Tag Type Analysis

The majority of VideoTag tags are single word, which is interesting as a conscious decision was made to not limit the format in which users could enter tags, as it was believed all tags regardless of format are useful at improving meta data for a video. Del.icio.us only allows single word tags as do other systems but some such as Last.fm allow multi word tags. It is interesting that the majority of users automatically tag as a single word and do not think to enter full phrase descriptions. It would be interesting to find out if experience at tagging affects the types of tags entered, with more experienced taggers using single word tags as pre-conditioned by systems like del.icio.us, and novice users entering a more varied range of single and multi word tags.

Table 2 compares the VideoTag and YouTube tags. Using this example the types of tag entered can be analysed in relation to the Golder and Huberman (2005) definitions of tag type. YouTube tags can be found to fall into the social tag functions of What or Who it is about (e.g. frog), and Qualities Or Characteristics (e.g. animation and funny) with funny being an Opinion Expression tag. VideoTag tags also have social tag functions and similarly to YouTube tags, are primarily What or Who it is about (e.g. frog, fly, two frogs eating flys) and Qualities Or Characteristics (e.g. cartoon, comedy, taunting, greedy). Few of the Qualities Or Characteristics tags were Opinion Expression tags. The majority of the tags describe the characters, objects or actions in the videos with a few Opinion Expression tags (e.g. funny, humour, silly). It is surprising that not more Opinion Expression tags were entered, they are particularly useful at categorising videos as well as formulating descriptions. It would be interesting to find out, in future research, whether the gameplay of VideoTag deterred users from entering opinion expression tags, by comparing the frequency of Opinion Expression tags to those in tagging systems such as del.icio.us. These results were general for all videos. This analysis implies that VideoTag managed to successfully encourage users to enter more descriptive tags for the videos.

This webpage has been created that shows thumbnails of each of the videos in VideoTag and lists the VideoTag tags generated as well as the original YouTube tags.

Table 2 Table comparing tags entered during the VideoTag experiment and YouTube tags for one example video from the VideoTag database.

References

BAR-ILAN, J., SHOHAM, S., IDAN, A., MILLER, Y. & SHACHAK, A. (2006) Structured vs. unstructured tagging ? A case study. Proceedings of the Collaborative Web Tagging Workshop (WWW ’06), Edinburgh, Scotland.

CATTUTO, C., LORETO, V. & PIETRONERO, L. (2007) Semiotic dynamics and collaborative tagging. Proceedings of the National Academy of Sciences (PNAS), 104(5), pp. 1461-1464.

FURNAS, G.W., LANDAUER, T.K., GOMEZ, L.M., DUMAIS, S. T. (1987) The vocabulary problem in human-system communication. Communications of the Association for Computing Machinery, 30(11), pp. 964-971.

GEISLER, G. and BURNS, S. (2007) Tagging video: conventions and strategies of the YouTube community. JCDL ’07: Proceedings of the 2007 conference on Digital libraries, pp. 480-480

GOLDER, S. & HUBERMAN, B. (2005) The Structure of Collaborative Tagging Systems [Online]. Available from:http://arxiv.org/abs/cs.DL/0508082 [cited 09-03-2007]

MARLOW, C., NAAMAN, M., BOYD, D. & DAVIS, M. (2006) HT06, tagging paper, taxonomy, Flickr, academic article, to read. HYPERTEXT ’06: Proceedings of the seventeenth conference on Hypertext and hypermedia, pp. 31-40.

Cognitive Level, Semantic Distance and Power Laws.

My brain hurts from reading around the subject of Cognitive Linguistics, power laws and Semantic Distance and trying to work out clearly in my mind how they are connected. In doing so i can better explain what I mean by improving quality of tags for video through VideoTag and make understood what I perceive to be a higher quality tag.

So here goes:

There are three cognitive levels of tags Superordinate, Basic and Subordinate. Basic level tags have the least cognitive cost to the user – that is they are thought of more quickly. They are more likely to have a high frequency as there is more likely to be agreement on Basic Level tags. Superordinate and subordinate have a higher cognitive cost. In relation to collaborative tagging – superordinate level is difficult to assess. It is most likely that the superordinate tag for videos is video and all basic and subordinate level tags then continue to categorise the video. When tagging a music video for instance, basic level tags may refer to musical genre e.g. rock, indie, dance but the tag music would also be a basic level tag rather than being a superordinate tag that defines the overall category for tags, because it defines the genre of the video.  Subordinate level tags on the other may reference the band name, more specific musical genres e.g. Techno, trance, emo, grunge, britpop etc. They may also name band members, cameo roles by celebrities in the video. characters in the video, define the narrative of the video and any specific actions. Keywords taken from the song lyrics would also be classed as subordinate level tags.

The tag cloud below is of My Chemical Romance’s tags on Last.fm – chosen because they have 2 of the most watched videos on YouTube. You Tube tags – my chemical romance famous last words (whilst I would categorise these tags as subordinate level based on the above definition, they also highlight how inadequate YouTube tags are at describing the videos.)

This helps to explain the power law of tags. Tags in the larger font (e.g. emo, rock, alternative) are basic level tags. Tags with smaller font are of subordinate level. In this instance the superordinate tag would be music, but as it is a music site all tags contained with in fall under the umbrella of the music superordinate tag. On a power law graph, the high frequency basic level tags would have high rank, the subordinate level low frequency tags will have low rank and appear in the long tail.

The 80/20 rule can be applied here, agreement of terms can be measured as being 20% based on the frequency of basic level tags. This leaves 80% of tags at subordinate level that describe the resource but may only be of relevance to a few users. In terms of building rich descriptions of video, these subordinate level tags are imperative as they go into more descriptive detail, have greater specificity and can provide a fuller picture as to what the video is about.

As for semantic distance – I have only recently started to read up on this so I am not 100% in my mind of the connection. I think that subordinate level tags are more likely to be semantically narrow because they are related by the basic level tag they are elaborating, making the basic level tags semantically broad.

So what is a high quality tag? In terms of improving descriptions of videos it is a tag of subordinate cognitive level, low rank and low frequency and is semantically narrow. It is worth mentioning though that subordinate level tags are only useful when placed in context with the basic level tag they are adding extra description too. So VideoTag needs to encourage both sets of tags to be useful as a tool to improve accessibility and search of video.

Why selecting videos for VideoTag is not as easy as you’d think.

I am warming up to working today – actually I’ve already written some notes and it’s only 10.00, so I thought I deserved a break. I have realised I have been neglecting the blogosphere and so subscribed to feeds from some of the most respected web2.0 blogs. Anyway looking at readwriteweb I saw this article, top 10 youtube videos of all time, which I found interesting as it’s written at the time when I had spent a month searching out videos for VideoTag and knew pretty much every popular video on there.

It sums up why I gave up on rss feeds to provide videos for VideoTag and why I cannot see a way that any version of the game would work synced directly to the YouTube api. The most watched/highly rated/most favourited videos are mostly music videos that are not going to benefit from the extra descriptions that tagging can provide as much as the amateur videos.

6 months on and the top 10 most watched of all time hasn’t really changed

1. Evolution of Dance

2. Avril Lavigne – Girlfriend

3. Lo que tú Quieras Oír

4. IMVU – http://www.IMVU.com

5. Timbaland – Apologize – Official Music Video

6. My Chemical Romance – Teenagers

7. My Chemical Romance – Famous Last Words

8. Timbaland – The Way I Are OFFICIAL MUSIC VIDEO

9. Akon – “Don’t Matter”

10. CANSEI DE SER SEXY – Music is My Hot Hot Sex

laughing baby has slipped to number 14.

Trying to harness people power and doing it successfully maybe proving people really will watch anything is some video about Britney Spears in a bikini. They want to be the most watched worst video of all time. Their intentions seem honourable, trying to rid the top 10 videos of music videos, but i couldn’t bring myself to watch anything about Britney once never mind 100 times a day.

And for the record, maybe it’s a British thing, but I didn’t even think the most watched video was that funny.

Contemplating the future

My first goal now that my Msc is finished is to write up the VideoTag experiment in a paper, in the hopes the research will get a bit more interest from academic circles.  My main hope is to secure funding for a PhD so i can continue researching tagging and progress VideoTag.

The main areas I next want to focus on are motivating users to tag and game motivation.  Whilst VideoTag has found that a game format can encourage users to tag, I want to find out what motivates users to play games, return to games and stay playing games for long periods.  As well as finding out more about why users tag and whether tagging behaviour is different depending on the resource.  I also want to investigate further whether blind, suggested or guided tagging generate tags of differing cognitive level and whether or not one method of tagging is more successful than another at generating more descriptive tags.  However, the VideoTag experiment highlighted that a game environment is not a good tool to use to produce the data to analyse these methods of tagging.  Therefore, a new experiment needs to be thought up – ideally analysing existing tag data, because to create a new experiment I will face the same user motivation problems.  However, it is rare for sites to use guided tagging so it could be hard to find existing data.  Things to think about.

As for VideoTag I want to redesign it give the homepage more impact, add in the golden tag idea, cut down the video length make a limit of 2 minutes.  Also maybe offer options to quit after 15 seconds and get a new video or continue watching.  It would also be good to add in a vote system at this point to rate the video boring or interesting.  I like the idea of using VideoTag data to potentially develop an interestingness filter for videos.  My long term aim with VideoTag is that by making improvements to the gameplay enough data will be generated that will then make it possible to experiment with using the data to improve accessibility and search of internet video. 

All Finished

Well I’m all finished.  Handed in my report, I’m pleased with all I’ve accomplished through this project.

Here is the Abstract for the report:

Through discussion and analysis of current research in collaborative tagging systems, an emerging area of research was discovered, improving accessibility and search of visual resources through tagging.  Of particular interest were two tagging projects ESP Game and Steve.Museum, where users were encouraged to tag images to improve accessibility and search of images.  VideoTag extends this research by harnessing the user motivations of Play and Competition to increase and improve the meta data of a selection of YouTube videos through tagging. 

The VideoTag tagging experiment consisted of a one player game where users were encouraged to tag a selection of sixty carefully chosen, funny or interesting YouTube Videos.  The videos were separated over five difficulty levels.  Gameplay was carefully planned in order to encourage users to tag the videos more descriptively, using tags of a subordinate rather than basic cognitive level.  The experiment was uncontrolled with random users being attracted to the game through promotion on various Web 2.0 sites.

Analysis of the results focused on whether a game environment is beneficial to encouraging users to tag videos.  Quantitative methods of analysis found VideoTag to be successful at increasing the amount of tags per video compared to YouTube.   A long tail effect was found to present in the tag data which allowed for qualitative analysis of the quality of the tags entered based on their cognitive level.  

As only a small selection of videos were used, tag data generated by the VideoTag experiment is not sufficient to test whether the data can improve search for those selected videos, or create descriptions to improve accessibility for visually impaired users.  Analysis and evaluation does discuss how VideoTag proves as a concept, game based tagging could be used to improve accessibility and search and there is scope for future research .