Global Video Game Sales and Reviews

Global Video Game Performance: Sales, Reviews, and Rankings

By Andy Bramwell [source]

About this dataset

The elements covered in this well-curated dataset include: The ranking of the game based on global sales under the column 'Rank'. This metric provides perspective on how popular or successful a particular game has been across countries in comparison to others during its time. Noting that video games' popularity could vary greatly from one geography to another due to factors like cultural nuances, gamer preferences, etc., regional sales have been marked separately for North America (North America), Europe (Europe), Japan (Japan) as well as for other parts of the World excluding these three regions under the column 'Rest of World'.

For easy identification among massive chunks of data, we've included each game's title (Game Title) along with additional categorization based on their genre (Genre). From action-packed adventures to strategic board-like scenarios or enchanted magic realms - classifications cover it all! In addition, detailed information about publishers can be found under 'Publisher', which grants insights about leading companies dominating market shares.

Further details expand into mentioning platforms such as PS4, Xbox, PC where these games can be played under 'Platform'. A unique attribute covered in this database is ‘Review’. Given that critique ratings play an influential role in engaging new players into trying out a particular video game or boosting existing user morale regarding their choice; this numeric representation ranging typically from 1-10 vividly captures public opinion about them.

Lastly, just for keeping tabs on ever-evolving gaming technology standards where newer versions often outshine predecessors irrespective of actual gameplay quality itself; having release years mentioned ('Year') proves beneficial for categorizing them chronologically. This helps correlate whether higher sales figures can sometimes merely be indicative of more people having access to necessary high-end gaming hardware during later periods.

In essence, this dataset titled ‘Video Games Sales.csv’ holds immense potential for informative deep-dives into the Video Game industry's trends and paradigms, forming a solid foundation for market research, academic purposes or personal projects

How to use the dataset

This dataset provides extensive information about various video game titles, their sales performance across multiple regions, publisher details and game reviews. Follow the steps outlined below to make the most out of this remarkable dataset!

1. Game Research & Evaluation:

With columns such as 'Game Title', 'Genre' and 'Review', you can research on particular games or genres that interest you. You can evaluate a game based on its review scores, delving into what makes a top-rated game.

2. Publisher Analysis:

The 'Publisher' column lets you track which publishers are behind the most successful games in terms of sales and reviews. This analysis could be useful for people interested in business trends in gaming industry or trying to identify potential innovative publishers.

3. Regional Market Trend Identification:

You can use data from columns like ‘North America’, ‘Europe’, ‘Japan’ and ‘Rest of World’ to study regional market trends for certain genres or platforms; it might enable one to recognize patterns over time or cultural preferences with regard to video games.

4. Global Sales Analysis:

Using the 'Global' column, you could observe which games have been globally successful, going beyond regional preferences by genre or platform.

5. Platform Insight:

The platform on which a particular game is available is another significant factor (e.g., PC, PS4, Xbox). By utilizing the data contained in this dataset regarding platforms, one may learn how platform choice impacts global sales as well as discern any correlation between preferred platform types among specific regions.

Remember that every statistical analysis begins with knowing your data - dive deep into each variable; explore patterns within variables before looking at correlations between different fields.

Don't forget - when engaged with comprehensive datasets like these - creativity is your only limit! Happy analyzing!

Research Ideas

• Trend Analysis: This dataset can be used to analyze the trends in video game preferences over the years based on genre, publisher, platform and region. It can provide interesting insights into how consumer tastes have evolved with time and which game genres are becoming more popular.
• Sales Forecasting: U...

This dataset contains a list of video games with sales greater than 100,000 copies. It was generated by a scrape of vgchartz.com.

Fields include

• Rank - Ranking of overall sales

• Name - The games name

• Platform - Platform of the games release (i.e. PC,PS4, etc.)

• Year - Year of the game's release

• Genre - Genre of the game

• Publisher - Publisher of the game

• NA_Sales - Sales in North America (in millions)

• EU_Sales - Sales in Europe (in millions)

• JP_Sales - Sales in Japan (in millions)

• Other_Sales - Sales in the rest of the world (in millions)

• Global_Sales - Total worldwide sales.

The script to scrape the data is available at https://github.com/GregorUT/vgchartzScrape. It is based on BeautifulSoup using Python. There are 16,598 records. 2 records were dropped due to incomplete information.

This is a simple survey dataset conducted within my computer science department, involving participation from over 100 students. The primary focus of the survey was to identify the reasons why computer science students engage in playing video games. The survey comprises only seven questions, including age, gender, whether the participant plays video games, student status, favorite game, most played game, and the most important reason for playing games. The findings reveal that a majority of participants indicated playing games for entertainment purposes.

In July 2025, total video games sales in the United States amounted to **** billion U.S. dollars, representing a five percent year-over-year increase. Generally speaking, the video game industry has its most important months in November and December, as video game software and hardware make very popular Christmas gifts. In December 2024, total U.S. video game sales surpassed **** billion U.S. dollars. Birth of the video game industry Although the largest regional market in terms of sales, as well as number of gamers, is Asia Pacific, the United States is also an important player within the global video games industry. In fact, many consider the United States as the birthplace of gaming as we know it today, fueled by the arcade game fever in the ’60s and the introduction of the first personal computers and home gaming consoles in the ‘70s. Furthermore, the children of those eras are the game developers and game players of today, the ones who have driven the movement for better software solutions, better graphics, better sound and more advanced interaction not only for video games, but also for computers and communication technologies of today. An ever-changing market However, the video game industry in the United States is not only growing, it is also changing in many ways. Due to increased internet accessibility and development of technologies, more and more players are switching from single-player console or PC video games towards multiplayer games, as well as social networking games and last, but not least, mobile games, which are gaining tremendous popularity around the world. This can be evidenced in the fact that mobile games accounted for ** percent of the revenue of the games market worldwide, ahead of both console games and downloaded or boxed PC games.

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F15126770%2F7d5374215511bb7cf264fab8a294bc3a%2Fheader.jpg?generation=1704969406449875&alt=media" alt="">

Data obtained using a program from the site backloggd.com.

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F15126770%2F9bb6a4f0ee6d69ea160b12f4d1ca3e30%2Fdata_1.jpg?generation=1704968884700538&alt=media" alt="">

About backloggd.com

"Backloggd is a place to keep your personal video game collection. Every game from every platform is here for you to log into your journal. Follow friends along the way to share your reviews and compare ratings. Then use filters to sort through your collection and see what matters to you. Keep a backlog of what you are currently playing and what you want to play, see the numbers change as you continue to log your playthroughs. There's Goodreads for books, Letterboxd for movies, and now Backloggd for games." - from the site backloggd.com.

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F15126770%2F4e12014a1f38e1167a5cf66202ebf9d7%2Fdata_2.jpg?generation=1704968935015630&alt=media" alt="">

"All game related metadata comes from the community driven database IGDB. This includes all game, company and platform data you see on the site." - from the site backloggd.com

https://www.googleapis.com/download/storage/v1/b/kaggle-user-content/o/inbox%2F15126770%2F799831cb18c3b74f1c3f6e8a023af723%2Fdata_3.jpg?generation=1704968996471054&alt=media" alt="">

What can you do with the data set?

If you are new to data analytics, try answering the following questions: - in what year did the active growth in the number of video games produced begin? What year was the most successful from this point of view? - on what day and month were the largest number of video games released? What could be the reason for this pattern? - is there a dependence of the rating of a video game on the number of reviews left or the total number of players? - which game genres, platforms and developers are the most common (the most video games released of all time)? - which game genres, platforms and developers have the highest total number of players (have the highest total number of players ever)? - which game genres, platforms and developers have the highest average video game ratings?

If you have enough experience, try solving a multi-label classification problem. Train a model that can classify a video game description into one or more genres: - which models are best suited for this, and which should not be used? - what is the best way to convert text to features? How will lemmatization of text affect the predictive ability of the model? - which metric should be chosen to evaluate the model? - Is the model calibrated enough after training to trust its probabilistic forecasts? - can adding new data improve the predictive ability of the model?

Field descriptions:

The data contains the following fields: 1. games - basic data: - id - video game identifier (primary key); - name - name of the video game; - date - release date of the video game; - rating - average rating of the video game; - reviews - number of reviews; - plays - total number of players; - playing - number of players currently; - backlogs - the number of additions of a video game to the backlog; - wishlists - the number of times a video game has been added to “favorites”; - description - description of the video game. 2. developers - developers (publishers): - id - video game identifier (foreign key); - developer - developer (publisher) of a video game. 3. platforms - gaming platforms: - id - video game identifier (foreign key); - platform - gaming platform. 4. genres - game genres: - id - video game identifier (foreign key); - genre - video game genre. 5. scores - user ratings: - id - video game identifier (foreign key); - score - score (from 0.5 to 5 in increments of 0.5); - amount - number of users. 6. Video game posters.

Data update

The website backloggd.com contains detailed roadmap with changes that may be implemented over time on the website, among them: - additional information about the game: DLC status, all companies, alternative names and other extensive information about the game; - categorization of games: which games are DLC, demo versions, canceled, beta versions, etc.; - personalized game covers: IGDB now supports localized covers; - release dates: games with one date are too easy, in this case, multiple release dates will be shown for different stages/regions.

!...

Cross-sectional survey of individuals aged 16–25 residing in Spain (n = 1,000), fielded in June 2025. The dataset captures demographics, gaming practices, exposure to gamblified mechanics (e.g., loot boxes, randomized rewards, time-limited offers), spending bands, emotional responses, and perceptions among non-players.The survey questionnaire (available as related material) consists of three sections: 1) sociodemographic data; 2) gamers (n=927); and 3) non-gamers (n=73). To distinguish between gamers and non-gamers, a filter question was placed at the end of section 1 (“Do you regularly play video games, mobile games, or digital games?”).The first section collects information on: sex/gender, municipality of residence, province/community of residence according to Nielsen area, size of municipality of residence, level of education completed, and current employment status.Block two collects information on estimated weekly hours of play, the device usually used, estimated monthly money spent on video games, and the two most played games. It also includes five-point Likert scale questions (disagree/agree) on issues related to random mechanisms, payments, or gaming habits.The third section repeats the same pattern of five-point Likert scale questions (disagree/agree) but from the perspective of non-video game players.

In 2024, Call of Duty: Black Ops 6, published by Activision Blizzard, was the top-selling video game in the United States based on dollar sales. EA Sports College Football 25 was in second place, followed by the shooter Helldivers II. The dominance of AAA gaming productions In the video-game industry, AAA (pronounced Triple-A) is a classification that is used when describing or talking about video games that are produced and distributed by major or mid-tier video game developers. These game productions typically have a large development and marketing budget. When looking at the top-selling video games in the United States in 2023, the majority of the best-selling titles were produced by the biggest gaming companies worldwide. In addition to several publicly listed video game companies, gaming platform owners Sony and Nintendo also produce top-selling games with Sony's Marvel's Spider-Man 2 and Nintendo’s The Legend of Zelda: Tears of the Kingdom also reaching rank four and five, respectively. Spotlight: Call of Duty series Activision Blizzard’s long-running Call of Duty franchise usually features in many end-of-the-year lists as it is one of the top-grossing and best-selling gaming franchises worldwide. As of April 2021, the Call of Duty (CoD) series has generated more than 400 million lifetime unit sales and the most recent entry in the series is Call of Duty: Modern Warfare III, which was released in November 2023. The annual releases of the game series are rotated between several game development studios which are subsidiaries of Activision Blizzard: Infinity Ward, Sledgehammer Games, Treyarch, and Raven Software.

Hundreds of millions of people play intellectually-demanding video games every day. What does individual performance on these games tell us about cognition? Here, we describe two studies that examine the potential link between intelligence and performance in one of the most popular video games genres in the world (Multiplayer Online Battle Arenas: MOBAs). In the first study, we show that performance in the popular MOBA League of Legends’ correlates with fluid intelligence as measured under controlled laboratory conditions. In the second study, we also show that the age profile of performance in the two most widely-played MOBAs (League of Legends and DOTA II) matches that of raw fluid intelligence. We discuss and extend previous videogame literature on intelligence and videogames and suggest that commercial video games can be useful as 'proxy' tests of cognitive performance at a global population level.

Context

This is a list of the most-played mobile games ordered by their player count, which includes downloads, registered accounts, and/or monthly active users.

Acknowledgements

Image: Photo by Howard Bouchevereau on Unsplash

The dataset has been obtained from Wikipedia

What if helping science simply means you play more video games? Talk about a win, win … and you don’t even have to move off the couch!

Nintendo Switch Dedicated Video Game Sales Units: 2017-2020

Quantifying Unit Sales Performance

By Charlie Hutcheson [source]

About this dataset

This dataset contains dedicated video game sales unit data for the Nintendo Switch platform from 2017 to 2020, as reported by official Nintendo Investor Relations. It provides an snapshot of consumers’ buying trends over the past four years and helps us gain insightful understanding into the introduction, expansion and success of this platform across global markets. The data can be used to analyze multiple aspects such as performance of specific titles/genres/franchises, changes in market expectations over time and more. This chart helps to visualize the dynamic changes in these sales units over that four-year timeframe. From this chart we can gain valuable understanding about how successful various releases have been on this gaming console, what titles drove its popularity levels and more useful insights that could help other developers in creating future products for similar platforms

More Datasets

For more datasets, click here.

Featured Notebooks

• 🚨 Your notebook can be here! 🚨!

How to use the dataset

This dataset contains historical sales unit data for the Nintendo Switch platform from 2017 to 2020. The original visualization provides a clear visual representation of the number of sales units over time. It is easy to discern which months have seen higher levels of sales, and which have seen lower ones.

Using this dataset, users can perform various analyses on the results to gain further insights into consumer trends and buy behavior associated with the Nintendo Switch platform. Users can also glean information on pricing strategies taken by Nintendo as well as consumer preferences over time in order to inform future business decisions.

In order to maximize use of this data set, users are encouraged to consider questions such as: What types of games do consumers prefer? How has their taste changed over time? What is the average amount spent per game by region or country? How often are certain consoles purchased or rented? And what role do discounts or promotions play in influencing purchasing decisions? By exploring these questions, users can begin understanding how different factors may be affecting overall demand for a product associated with the Nintendo Switch platform.

By analyzing this dataset, users also get an insight into how other competitors within the industry are affecting sales performance and allowing them take steps necessary for either surpassing competitors or maintaining dominance through suitable tactics like improved marketing campaigns or better-priced products that appeal more strongly customers’ needs and wants . In addition, examining this data enables companies keenly understand customer demands at detailed levels including whether customers prefer switch game bundles with extra features like custom skins etc., titles released during special times such as a holiday season that incite strong demand among buyers and also relevant discounts/promotions offered during times when people want/needing much needed break from regular routine life.. Ultimately , gaining greater insight into customer objectives allows firms efficiently manage their costs while maximizing profits through effective decisions based on reliable datasets such as that contained in this one instead rarely updated manual counts/observations which dont just lack comprehensiveness but also accuracy in nature.

Research Ideas

• Producing a mobile application to present the sales units in an intuitive and interactive way;
• Utilizing the data for machine learning algorithms to predict and analyze trends in Nintendo Switch dedicated video game sales units;
• Creating infographics and visualizations that can be used for promotional materials or to educate customers about the success of Nintendo Switch dedicated game sales

Acknowledgements

If you use this dataset in your research, please credit the original authors. Data Source

License

License: Dataset copyright by authors - You are free to: - Share - copy and redistribute the material in any medium or format for any purpose, even commercially. - Adapt - remix, transform, and build upon the material for any purpose, even commercially. - You must: - Give appropriate credit - Provide a link to the license, and indicate if changes were made. - ShareAlike - You must distribute your contributions under the same license as the or...

Here are a few use cases for this project:

1. Sports Training Analysis: Used by sports coaches or trainers for analyzing player roles in sports like football, soccer, or rugby. The model can identify player classes and enhance strategies in training or during live games.

2. Gaming and E-Sports: The model can be used in the design and development of sports video games, helping to accurately identify player classes, create realistic AI competitors and create more immersive experiences.

3. Physical Education and Fitness: Could be used in fitness apps or PE classes to record and assist in team-based exercises, helping coaches or teachers better organize games and understand playing styles of individuals.

4. Event Organizing: Organizers of field events or team-building activities could use this model to better understand participant roles and interactions, improving future event planning.

5. Surveillance and Security: In scenarios like public parks or stadiums, where groups of people often play sports, this model could be used in surveillance systems to observe and differentiate between different types of players for crowd control or safety incidents.

A global consumer survey conducted in March 2024 found that 18 percent of respondents were more likely to buy a video game if it was advertised as a collector or limited edition. However, 45 percent of respondents stated that they were not interested in limited edition releases.

Here are a few use cases for this project:

1. Video Game Development: This model could be used to test and optimize game environments, identifying the counts and types of objects present at any given time. It could enable developers to balance gameplay by analyzing the occurrences of various in-game objects.

2. AI Gaming Assistant: "Yolo Labeling 2nd" can help build an AI gaming assistant that can provide real-time guidance to players. The assistant can predict potential threats based on the recognition of objects such as enemies or weapons.

3. Moderation in Gaming Platforms: The model can be used as a tool for moderating online gaming platforms by identifying potential violent content. If a user shares a video or an image of the game showing too many elements like Guns, Knives, Friendly-Dead or Enemy-Dead, it can flag them to be reviewed by human moderators.

4. Game Streaming Analysis: Streamers who are playing games live could use this model to analyze gameplay and provide real-time stats to their viewers. It could track the numbers and types of certain in-game objects, adding an extra layer of interaction during the stream.

5. Video Game Reviews: Reviewers could use this model to analyze the content of games, identifying the balance between non-violent and violent elements while making comparisons across different games, thus providing more comprehensible content analysis in their reviews.

This dataset presents a comprehensive and structured collection of video game metadata and user reviews from the Steam platform, covering the period between January 2020 and December 2024. It was compiled to support research into how various game attributes influence user satisfaction, engagement, and review behavior. The central research hypothesis behind this work suggests that specific characteristics of video games, such as genre, pricing, and supported platform, are closely associated with trends in user sentiment and review volume. Understanding these patterns can contribute to predictive models of game reception and improve design and marketing strategies for future releases.

To explore this hypothesis, data was gathered in two phases. In the first phase, metadata for all games listed on Steam during the target period was collected using the official Steam API. Each game was identified by its unique AppID and evaluated to ensure data completeness. The scraper retrieved details including the game title, release date, genres, supported languages, age restrictions, and pricing information. Games that were unreleased or launched before 2020 were excluded from the dataset. This resulted in a refined metadata file, stored as games.json, containing detailed information on 23,107 Steam games released from 2020 onward.

In the second phase, a dedicated script was used to collect user reviews for each game in the metadata file. The review collection process filtered out games with fewer than 25 reviews to avoid bias due to insufficient data. For the remaining games, reviews were gathered in all available languages to ensure a culturally diverse and inclusive dataset. Reviews were saved in individual CSV files named using the game’s AppID and the number of reviews it contains. Each file includes structured rows with fields such as review text, language, rating, and possibly vote counts depending on the API’s response format. This resulted in over 31 million reviews across more than 23,000 games, forming a robust basis for textual and quantitative analysis.

The data reveals several meaningful trends. Free-to-play games tend to attract higher review volumes, although not necessarily higher user ratings. Games within specific genres, such as role-playing, simulation, and survival, often have longer and more detailed reviews, indicating deeper user engagement.

By releasing both the metadata and reviews together, this dataset offers a multidimensional view of the Steam game landscape from 2020 to 2024. It captures user engagement during a pivotal period in digital gaming and provides a foundation for future research in user behavior, content personalization, and the evolving dynamics of online platforms.

Comprehensive ranking dataset of the top 100 YouTube channels in the Gaming category. This dataset features 100 channels with detailed statistics including subscriber counts, total video views, video count, and global rankings. The leading channel has 110,000,000 subscribers and 29,436,109,895 total views. Each entry includes comprehensive metrics to analyze channel performance, growth trends, and competitive positioning. This dataset is regularly updated to reflect the latest YouTube channel statistics and ranking changes, providing valuable insights for content creators, marketers, and researchers analyzing YouTube ecosystem trends and channel performance benchmarks.

Version 4 of the dataset is available (Sep 19 2019)!

Note this version has significantly more data than Version 2.

Dataset description paper (full version) is available!

https://arxiv.org/pdf/1903.06754.pdf (updated Sep 7 2019)

Tools for visualizing the data is available!

https://github.com/corgiTrax/Gaze-Data-Processor

=========================== Dataset Description ===========================

We provide a large-scale, high-quality dataset of human actions with simultaneously recorded eye movements while humans play Atari video games. The dataset consists of 117 hours of gameplay data from a diverse set of 20 games, with 8 million action demonstrations and 328 million gaze samples. We introduce a novel form of gameplay, in which the human plays in a semi-frame-by-frame manner. This leads to near-optimal game decisions and game scores that are comparable or better than known human records. For every game frame, its corresponding image frame, the human keystroke action, the reaction time to make that action, the gaze positions, and immediate reward returned by the environment were recorded.

Q & A: Why frame-by-frame game mode?

Resolving state-action mismatch: Closed-loop human visuomotor reaction time is around 250-300 milliseconds. Therefore, during gameplay, state (image) and action that are simultaneously recorded at time step t could be mismatched. Action at time t could be intended for a state 250-300ms ago. This effect causes a serious issue for supervised learning algorithms, since label at and input st are no longer matched. Frame-by-frame game play ensures states and actions are matched at every timestep.

Maximizing human performance: Frame-by-frame mode makes gameplay more relaxing and reduces fatigue, which could normally result in blinking and would corrupt eye-tracking data. More importantly, this design reduces sub-optimal decisions caused by inattentive blindness.

Highlighting critical states that require multiple eye movements: Human decision time and all eye movements were recorded at every frame. The states that could lead to a large reward or penalty, or the ones that require sophisticated planning, will take longer and require multiple eye movements for the player to make a decision. Stopping gameplay means that the observer can use eye-movements to resolve complex situations. This is important because if the algorithm is going to learn from eye-movements it must contain all “relevant” eye-movements.

============================ Readme ============================

1. meta_data.csv: meta data for the dataset., including:

GameName: String. Game name. e.g., “alien” indicates the trial is collected for game Alien (15 min time limit). “alien_highscore” is the trajectory collected from the best player’s highest score (2 hour limit). See dataset description paper for details.

trial_id: Integer. One can use this number to locate the associated .tar.bz2 file and label file.

subject_id: Char. Human subject identifiers.

load_trial: Integer. 0 indicates that the game starts from scratch. If this field is non-zero, it means that the current trial continues from a saved trial. The number indicates the trial number to look for.

highest_score: Integer. The highest game score obtained from this trial.

total_frame: Number of image frames in the .tar.bz2 repository.

total_game_play_time: Integer. game time in ms.

total_episode: Integer. number of episodes in the current trial. An episode terminates when all lives are consumed.

avg_error: Float. Average eye-tracking validation error at the end of each trial in visual degree (1 visual degree = 1.44 cm in our experiment). See our paper for the calibration/validation process.

max_error: Float. Max eye-tracking validation error.

low_sample_rate: Percentage. Percentage of frames with less than 10 gaze samples. The most common reason for this is blinking.

frame_averaging: Boolean. The game engine allows one to turn this on or off. When turning on (TRUE), two consecutive frames are averaged, this alleviates screen flickering in some games.

fps: Integer. Frame per second when an action key is held down.

1. [game_name].zip files: these include data for each game, including:

*.tar.bz2 files: contains game image frames. The filename indicates its trial number.

*.txt files: label file for each trial, including:

frame_id: String. The ID of a frame, can be used to locate the corresponding image frame in .tar.bz2 file.

episode_id: Integer (not available for some trials). Episode number, starting from 0 for each trial. A trial could contain a single trial or multiple trials.

score: Integer (not available for some trials). Current game score for that frame.

duration(ms): Integer. Time elapsed until the human player made a decision.

unclipped_reward: Integer. Immediate reward returned by the game engine.

action: Integer. See action_enums.txt for the mapping. This is consistent with the Arcade Learning Environment setup.

gaze_positions: Null/A list of integers: x0,y0,x1,y1,...,xn,yn. Gaze positions for the current frame. Could be null if no gaze. (0,0) is the top-left corner. x: horizontal axis. y: vertical.

1. action_enums.txt: contains integer to action mapping defined by the Arcade Learning Environment.

============================ Citation ============================

If you use the Atari-HEAD in your research, we ask that you please cite the following:

@misc{zhang2019atarihead,

title={Atari-HEAD: Atari Human Eye-Tracking and Demonstration Dataset},


author={Ruohan Zhang and Calen Walshe and Zhuode Liu and Lin Guan and Karl S. Muller and Jake A. Whritner and Luxin Zhang and Mary M. Hayhoe and Dana H. Ballard},


year={2019},


eprint={1903.06754},


archivePrefix={arXiv},


primaryClass={cs.LG}

}

Zhang, Ruohan, Zhuode Liu, Luxin Zhang, Jake A. Whritner, Karl S. Muller, Mary M. Hayhoe, and Dana H. Ballard. "AGIL: Learning attention from human for visuomotor tasks." In Proceedings of the European Conference on Computer Vision (ECCV), pp. 663-679. 2018.

@inproceedings{zhang2018agil,

title={AGIL: Learning attention from human for visuomotor tasks},

author={Zhang, Ruohan and Liu, Zhuode and Zhang, Luxin and Whritner, Jake A and Muller, Karl S and Hayhoe, Mary M and Ballard, Dana H},

booktitle={Proceedings of the European Conference on Computer Vision (ECCV)},

pages={663--679},

year={2018}

}

The purpose of this project is to determine how college students distribute their activities in time (with a particular focus on academic and athletic activities) and to examine the factors that influence such distributions.Each R reported once about each of the seven days of the week and an additional time about either Saturday or Sunday. Rs were told the week before they were to report which day was assigned and were given a report form to complete during that day. They entered the i nformation from that form when they returned the next week.The activity codes included were: 0: Sleeping. 1: Attending classes. 2: Studying or preparing classroom assignments. 3: Working at a jog (including CAPS). 4: Cooking, home chores, laundry, grocery shopping. 5: Errands, non-grocery shopping, gardening, animal care. 6: Eating. 7: Bathing, getting dressed, etc. 8: Sports, exercising, other physical activities. 9: Playing competitive games (cards, darts, videogames, frisbee, chess, Tr ivial Pursuit, etc.). 10: Participating in UNC-sponsored organizations (student government, band, sorority, etc.). 11: Listening to the radio. 12: Watching TV. 13: Reading for pleasure (not studying or reading for class). 14: Going to a movie. 15: Attending a cultural event (such as a play, concert, or museum). 16: Attending a sports event as a spectator. 17: Partying. 18: Religious activities. 19: Conversation. 20: Travel. 21: Resting. 22: Doing other things DIARY1-8: These datasets contain a matrix of activities by times for a particular day. Included is time period, activity code (see above), # of friends present, # of others present. (Rs were allowed to report doing two activities at once. In these cases they were also asked to report the % of time during the time period affected which was allocated to the first of the two activities listed.)THE DIARY DATASETS ARE STORED IN RAW FORM. SUMMARY FILES, CALLED TIMEREP, CONTAIN MOST SUMMA RY INFORMATION WHICH MIGHT BE USED IN ANALYSES. THE DIARY DATASETS CAN BE LISTED TO ALLOW UNIQUE CODING OF THE ORIGINAL DATA. Each R reported once about each of the seven days of the week and an additional time about either Saturday or Sunday.TIMEREP: The TIMEREP dataset is a summary file which gives the amount of time spent on each activity during each of the eight reporting periods and also includes more detailed information about many of the activities from follow-up questions which were asked if the respondent reported having engaged in certain activities. Data from additional questions asked of every respondent after each diary entry are also included: contact with family members, number of alcoholic drinks consumed during the 24 hour period reported on, number of friends and others present while drinking, number of cigarettes smoked on day reported about, and number of classes skipped on day reported about. Follow-up questions include detail about kind of physical activity or sports participation, kind of university organization, kind of radio program listened to and place of listening, kind of TV program watched and place of watching, kind of reading material read and topic, alcohol consumed while partying and place of partying, conversation topics, kind of travel, activities included in 'other' category.Special processing is required to put the dataset into SAS format. See spec for details.

Here are a few use cases for this project:

1. Video Game Development: Developers can use the "orange_bots" model to create more immersive and interactive games, especially for games that include 'bots' as part of their character roster. The model could help developers more easily create non-player characters (NPCs) and categorize them properly.

2. eSports Analysis: This model could be used to study and analyze gameplay in eSports, particularly in recognizing bot strategies and player interactions with bots. This data could then be used to improve game design, player training, or competitive strategies.

3. Content Moderation: For platforms hosting user-generated gaming content, the model can help identify the portions of the game that include bots. This can assist in moderating content, ensuring the fair play principles are adhered to, and identifying any bot-related cheating.

4. User-Generated Content Curation: The model can be used as a tool for curating user-generated content, like videos or streamed content featuring gameplay. By recognizing bots, videos could be correctly labeled and categorized for easier discovery.

5. Interactive Entertainment: This model could be employed in theme parks or virtual reality experiences for user interaction with bots. As users engage with virtual bots, their behaviors and responses can be analyzed to enhance the user experience.

Here are a few use cases for this project:

1. Sports Training: This model can be utilized to analyze movements in golf swings and shots. It can track and study the trajectory and speed of a golf ball after a shot or during practice sessions, providing useful feedback to players regarding their technique.

2. Recreation Management: Golf courses could use this model to monitor each playing hole, ensuring balls are located in the right areas and detect/or anticipate any golf course maintenance issues, such as balls continuously going out of bounds.

3. Golf Broadcasting and Media: The GolfProject model can support real-time sportscast applications. It would assist in precisely tracing the golf ball, enhancing viewer experiences by offering detailed insights about each shot.

4. Golf Equipment Development: Companies designing golf balls or pins could use this model for quality control, ensuring products have the accurate aspects for detection or comparing how their ball or pin appears compared to those of other competitors.

5. Game Development: Developers of golf-based video games and virtual reality experiences can use this model to create more realistic golf physics and visuals, improving user experiences.