Center for Machine Learning and Intelligent Systems
About  Citation Policy  Donate a Data Set  Contact

Repository Web            Google
View ALL Data Sets

Browse Through:

Default Task - Undo

Classification (129)
Regression (37)
Clustering (30)
Other (17)

Attribute Type

Categorical (8)
Numerical (97)
Mixed (13)

Data Type

Multivariate (109)
Univariate (6)
Sequential (19)
Time-Series (30)
Text (13)
Domain-Theory (5)
Other (2)


Life Sciences (26)
Physical Sciences (16)
CS / Engineering (49)
Social Sciences (6)
Business (8)
Game (4)
Other (19)

# Attributes

Less than 10 (25)
10 to 100 (68)
Greater than 100 (29)

# Instances - Undo

Less than 100 (11)
100 to 1000 (103)
Greater than 1000 (129)

Format Type

Matrix (101)
Non-Matrix (28)

129 Data Sets

Table View  List View

1. Heterogeneity Activity Recognition: The Heterogeneity Human Activity Recognition (HHAR) dataset from Smartphones and Smartwatches is a dataset devised to benchmark human activity recognition algorithms (classification, automatic data segmentation, sensor fusion, feature extraction, etc.) in real-world contexts; specifically, the dataset is gathered with a variety of different device models and use-scenarios, in order to reflect sensing heterogeneities to be expected in real deployments.

2. HIGGS: This is a classification problem to distinguish between a signal process which produces Higgs bosons and a background process which does not.

3. HEPMASS: The search for exotic particles requires sorting through a large number of collisions to find the events of interest. This data set challenges one to detect a new particle of unknown mass.

4. Record Linkage Comparison Patterns: Element-wise comparison of records with personal data from a record linkage setting. The task is to decide from a comparison pattern whether the underlying records belong to one person.

5. SUSY: This is a classification problem to distinguish between a signal process which produces supersymmetric particles and a background process which does not.

6. Gas sensor array under dynamic gas mixtures: The data set contains the recordings of 16 chemical sensors exposed to two dynamic gas mixtures at varying concentrations. For each mixture, signals were acquired continuously during 12 hours.

7. KDD Cup 1999 Data: This is the data set used for The Third International Knowledge Discovery and Data Mining Tools Competition, which was held in conjunction with KDD-99

8. PAMAP2 Physical Activity Monitoring: The PAMAP2 Physical Activity Monitoring dataset contains data of 18 different physical activities, performed by 9 subjects wearing 3 inertial measurement units and a heart rate monitor.

9. URL Reputation: Anonymized 120-day subset of the ICML-09 URL data containing 2.4 million examples and 3.2 million features.

10. YouTube Comedy Slam Preference Data: This dataset provides user vote data on which video from a pair of videos is funnier collected on YouTube Comedy Slam. The task is to automatically predict this preference based on video metadata.

11. Poker Hand: Purpose is to predict poker hands

12. Covertype: Forest CoverType dataset

13. Online Retail: This is a transnational data set which contains all the transactions occurring between 01/12/2010 and 09/12/2011 for a UK-based and registered non-store online retail.

14. News Aggregator: References to news pages collected from an web aggregator in the period from 10-March-2014 to 10-August-2014. The resources are grouped into clusters that represent pages discussing the same story.

15. Machine Learning based ZZAlpha Ltd. Stock Recommendations 2012-2014: The data here are the ZZAlpha® machine learning recommendations made for various US traded stock portfolios the morning of each day during the 3 year period Jan 1, 2012 - Dec 31, 2014.

16. Census-Income (KDD): This data set contains weighted census data extracted from the 1994 and 1995 current population surveys conducted by the U.S. Census Bureau.

17. Skin Segmentation: The Skin Segmentation dataset is constructed over B, G, R color space. Skin and Nonskin dataset is generated using skin textures from face images of diversity of age, gender, and race people.

18. Educational Process Mining (EPM): A Learning Analytics Data Set: Educational Process Mining data set is built from the recordings of 115 subjects' activities through a logging application while learning with an educational simulator.

19. microblogPCU: MicroblogPCU data is crawled from sina weibo microblog[]. This data can be used to study machine learning methods as well as do some social network research.

20. Wearable Computing: Classification of Body Postures and Movements (PUC-Rio): A dataset with 5 classes (sitting-down, standing-up, standing, walking, and sitting) collected on 8 hours of activities of 4 healthy subjects. We also established a baseline performance index.

21. Localization Data for Person Activity: Data contains recordings of five people performing different activities. Each person wore four sensors (tags) while performing the same scenario five times.

22. Buzz in social media : This data-set contains examples of buzz events from two different social networks: Twitter, and Tom's Hardware, a forum network focusing on new technology with more conservative dynamics.

23. MiniBooNE particle identification: This dataset is taken from the MiniBooNE experiment and is used to distinguish electron neutrinos (signal) from muon neutrinos (background).

24. TV News Channel Commercial Detection Dataset: TV Commercials data set consists of standard audio-visual features of video shots extracted from 150 hours of TV news broadcast of 3 Indian and 2 international news channels ( 30 Hours each).

25. YouTube Multiview Video Games Dataset: This dataset contains about 120k instances, each described by 13 feature types, with class information, specially useful for exploring multiview topics (cotraining, ensembles, clustering,..).

26. Reuters RCV1 RCV2 Multilingual, Multiview Text Categorization Test collection: This test collection contains feature characteristics of documents originally written in five different languages and their translations, over a common set of 6 categories.

27. Diabetes 130-US hospitals for years 1999-2008: This data has been prepared to analyze factors related to readmission as well as other outcomes pertaining to patients with diabetes.

28. Connect-4: Contains connect-4 positions

29. KEGG Metabolic Reaction Network (Undirected): KEGG Metabolic pathways modeled as un-directed reaction network. Variety of graphical features presented.

30. Dataset for Sensorless Drive Diagnosis: Features are extracted from motor current. The motor has intact and defective components. This results in 11 different classes with different conditions.

31. Statlog (Shuttle): The shuttle dataset contains 9 attributes all of which are numerical. Approximately 80% of the data belongs to class 1

32. KEGG Metabolic Relation Network (Directed): KEGG Metabolic pathways modeled as directed relation network. Variety of graphical features presented.

33. Adult: Predict whether income exceeds $50K/yr based on census data. Also known as "Census Income" dataset.

34. Census Income: Predict whether income exceeds $50K/yr based on census data. Also known as "Adult" dataset.

35. Tamilnadu Electricity Board Hourly Readings: This data can be effectively produced the result to fewer parameter of the Load profile can be reduced in the Database

36. Bank Marketing: The data is related with direct marketing campaigns (phone calls) of a Portuguese banking institution. The classification goal is to predict if the client will subscribe a term deposit (variable y).

37. UJIIndoorLoc-Mag: The UJIIndoorLoc-Mag is an indoor localization database to test Indoor Positioning System that rely on Earth's magnetic field variations.

38. Online News Popularity: This dataset summarizes a heterogeneous set of features about articles published by Mashable in a period of two years. The goal is to predict the number of shares in social networks (popularity).

39. Weight Lifting Exercises monitored with Inertial Measurement Units: Six young health subjects were asked to perform 5 variations of the biceps curl weight lifting exercise. One of the variations is the one predicted by the health professional.

40. Nomao: Nomao collects data about places (name, phone, localization...) from many sources. Deduplication consists in detecting what data refer to the same place. Instances in the dataset compare 2 spots.

41. default of credit card clients: This research aimed at the case of customers’ default payments in Taiwan and compares the predictive accuracy of probability of default among six data mining methods.

42. Chess (King-Rook vs. King): Chess Endgame Database for White King and Rook against Black King (KRK).

43. Grammatical Facial Expressions: This dataset supports the development of models that make possible to interpret Grammatical Facial Expressions from Brazilian Sign Language (Libras).

44. Reuters-21578 Text Categorization Collection: This is a collection of documents that appeared on Reuters newswire in 1987. The documents were assembled and indexed with categories.

45. UJIIndoorLoc: The UJIIndoorLoc is a Multi-Building Multi-Floor indoor localization database to test Indoor Positioning System that rely on WLAN/WiFi fingerprint.

46. Occupancy Detection : Experimental data used for binary classification (room occupancy) from Temperature,Humidity,Light and CO2. Ground-truth occupancy was obtained from time stamped pictures that were taken every minute.

47. Letter Recognition: Database of character image features; try to identify the letter

48. MAGIC Gamma Telescope: Data are MC generated to simulate registration of high energy gamma particles in an atmospheric Cherenkov telescope

49. Gas sensor arrays in open sampling settings: The dataset contains 18000 time-series recordings from a chemical detection platform at six different locations in a wind tunnel facility in response to ten high-priority chemical gaseous substances

50. p53 Mutants: The goal is to model mutant p53 transcriptional activity (active vs inactive) based on data extracted from biophysical simulations.

51. EEG Eye State: The data set consists of 14 EEG values and a value indicating the eye state.

52. Gas Sensor Array Drift Dataset at Different Concentrations: This archive contains 13910 measurements from 16 chemical sensors exposed to 6 different gases at various concentration levels.

53. Gas Sensor Array Drift Dataset: This archive contains 13910 measurements from 16 chemical sensors utilized in simulations for drift compensation in a discrimination task of 6 gases at various levels of concentrations.

54. Gisette: GISETTE is a handwritten digit recognition problem. The problem is to separate the highly confusible digits '4' and '9'. This dataset is one of five datasets of the NIPS 2003 feature selection challenge.

55. Indoor User Movement Prediction from RSS data: This dataset contains temporal data from a Wireless Sensor Network deployed in real-world office environments. The task is intended as real-life benchmark in the area of Ambient Assisted Living.

56. Nursery: Nursery Database was derived from a hierarchical decision model originally developed to rank applications for nursery schools.

57. Cuff-Less Blood Pressure Estimation: This Data set provides preprocessed and cleaned vital signals which can be used in designing algorithms for cuff-less estimation of the blood pressure.

58. UJI Pen Characters (Version 2): A pen-based database with more than 11k isolated handwritten characters

59. Pen-Based Recognition of Handwritten Digits: Digit database of 250 samples from 44 writers

60. Smartphone-Based Recognition of Human Activities and Postural Transitions: Activity recognition data set built from the recordings of 30 subjects performing basic activities and postural transitions while carrying a waist-mounted smartphone with embedded inertial sensors.

61. Firm-Teacher_Clave-Direction_Classification: The data are binary attack-point vectors and their clave-direction class(es) according to the partido-alto-based paradigm.

62. Human Activity Recognition Using Smartphones: Human Activity Recognition database built from the recordings of 30 subjects performing activities of daily living (ADL) while carrying a waist-mounted smartphone with embedded inertial sensors.

63. EMG Physical Action Data Set: The Physical Action Data Set includes 10 normal and 10 aggressive physical actions that measure the human activity. The data have been collected by 4 subjects using the Delsys EMG wireless apparatus.

64. Gesture Phase Segmentation: The dataset is composed by features extracted from 7 videos with people gesticulating, aiming at studying Gesture Phase Segmentation. It contains 50 attributes divided into two files for each video.

65. Daily and Sports Activities: The dataset comprises motion sensor data of 19 daily and sports activities each performed by 8 subjects in their own style for 5 minutes. Five Xsens MTx units are used on the torso, arms, and legs.

66. Spoken Arabic Digit: This dataset contains timeseries of mel-frequency cepstrum coefficients (MFCCs) corresponding to spoken Arabic digits. Includes data from 44 male and 44 female native Arabic speakers.

67. Online Handwritten Assamese Characters Dataset: This is a dataset of 8235 online handwritten assamese characters. The “online” process involves capturing of data as text is written on a digitizing tablet with an electronic pen.

68. Mushroom: From Audobon Society Field Guide; mushrooms described in terms of physical characteristics; classification: poisonous or edible

69. ISOLET: Goal: Predict which letter-name was spoken--a simple classification task.

70. Thyroid Disease: 10 separate databases from Garavan Institute

71. Australian Sign Language signs: This data consists of sample of Auslan (Australian Sign Language) signs. Examples of 95 signs were collected from five signers with a total of 6650 sign samples.

72. Musk (Version 2): The goal is to learn to predict whether new molecules will be musks or non-musks

73. HIV-1 protease cleavage: The data contains lists of octamers (8 amino acids) and a flag (-1 or 1) depending on whether HIV-1 protease will cleave in the central position (between amino acids 4 and 5).

74. Statlog (Landsat Satellite): Multi-spectral values of pixels in 3x3 neighbourhoods in a satellite image, and the classification associated with the central pixel in each neighbourhood

75. First-order theorem proving: Given a theorem, predict which of five heuristics will give the fastest proof when used by a first-order prover. A sixth prediction declines to attempt a proof, should the theorem be too difficult.

76. Artificial Characters: Dataset artificially generated by using first order theory which describes structure of ten capital letters of English alphabet

77. Turkiye Student Evaluation: This data set contains a total 5820 evaluation scores provided by students from Gazi University in Ankara (Turkey). There is a total of 28 course specific questions and additional 5 attributes.

78. Bach Choral Harmony: The data set is composed of 60 chorales (5665 events) by J.S. Bach (1675-1750). Each event of each chorale is labelled using 1 among 101 chord labels and described through 14 features.

79. Optical Recognition of Handwritten Digits: Two versions of this database available; see folder

80. SMS Spam Collection: The SMS Spam Collection is a public set of SMS labeled messages that have been collected for mobile phone spam research.

81. Page Blocks Classification: The problem consists of classifying all the blocks of the page layout of a document that has been detected by a segmentation process.

82. Wall-Following Robot Navigation Data: The data were collected as the SCITOS G5 robot navigates through the room following the wall in a clockwise direction, for 4 rounds, using 24 ultrasound sensors arranged circularly around its 'waist'.

83. Waveform Database Generator (Version 1): CART book's waveform domains

84. Waveform Database Generator (Version 2): CART book's waveform domains

85. Wine Quality: Two datasets are included, related to red and white vinho verde wine samples, from the north of Portugal. The goal is to model wine quality based on physicochemical tests (see [Cortez et al., 2009],

86. Wilt: High-resolution Remote Sensing data set (Quickbird). Small number of training samples of diseased trees, large number for other land cover. Testing data set from stratified random sample of image.

87. Spambase: Classifying Email as Spam or Non-Spam

88. Madelon: MADELON is an artificial dataset, which was part of the NIPS 2003 feature selection challenge. This is a two-class classification problem with continuous input variables. The difficulty is that the problem is multivariate and highly non-linear.

89. Abalone: Predict the age of abalone from physical measurements

90. Farm Ads: This data was collected from text ads found on twelve websites that deal with various farm animal related topics. The binary labels are based on whether or not the content owner approves of the ad.

91. Internet Advertisements: This dataset represents a set of possible advertisements on Internet pages.

92. Chess (King-Rook vs. King-Pawn): King+Rook versus King+Pawn on a7 (usually abbreviated KRKPA7).

93. Molecular Biology (Splice-junction Gene Sequences): Primate splice-junction gene sequences (DNA) with associated imperfect domain theory

94. sEMG for Basic Hand movements: The “sEMG for Basic Hand movements” includes 2 databases of surface electromyographic signals of 6 hand movements using Delsys' EMG System. Healthy subjects conducted six daily life grasps.

95. Sentiment Labelled Sentences: The dataset contains sentences labelled with positive or negative sentiment.

96. Vicon Physical Action Data Set: The Physical Action Data Set includes 10 normal and 10 aggressive physical actions that measure the human activity. The data have been collected by 10 subjects using the Vicon 3D tracker.

97. Character Trajectories: Multiple, labelled samples of pen tip trajectories recorded whilst writing individual characters. All samples are from the same writer, for the purposes of primitive extraction. Only characters with a single pen-down segment were considered.

98. Activities of Daily Living (ADLs) Recognition Using Binary Sensors: This dataset comprises information regarding the ADLs performed by two users on a daily basis in their own homes.

99. Dexter: DEXTER is a text classification problem in a bag-of-word representation. This is a two-class classification problem with sparse continuous input variables. This dataset is one of five datasets of the NIPS 2003 feature selection challenge.

100. seismic-bumps: The data describe the problem of high energy (higher than 10^4 J) seismic bumps forecasting in a coal mine. Data come from two of longwalls located in a Polish coal mine.

101. Australian Sign Language signs (High Quality): This data consists of sample of Auslan (Australian Sign Language) signs. 27 examples of each of 95 Auslan signs were captured from a native signer using high-quality position trackers

102. OPPORTUNITY Activity Recognition: The OPPORTUNITY Dataset for Human Activity Recognition from Wearable, Object, and Ambient Sensors is a dataset devised to benchmark human activity recognition algorithms (classification, automatic data segmentation, sensor fusion, feature extraction, etc).

103. Ozone Level Detection: Two ground ozone level data sets are included in this collection. One is the eight hour peak set (, the other is the one hour peak set ( Those data were collected from 1998 to 2004 at the Houston, Galveston and Brazoria area.

104. Reuter_50_50: The dataset is used for authorship identification in online Writeprint which is a new research field of pattern recognition.

105. Phishing Websites: This dataset collected mainly from: PhishTank archive, MillerSmiles archive, Google’s searching operators.

106. Image Segmentation: Image data described by high-level numeric-valued attributes, 7 classes

107. Statlog (Image Segmentation): This dataset is an image segmentation database similar to a database already present in the repository (Image segmentation database) but in a slightly different form.

108. Cardiotocography: The dataset consists of measurements of fetal heart rate (FHR) and uterine contraction (UC) features on cardiotocograms classified by expert obstetricians.

109. Twitter Data set for Arabic Sentiment Analysis: This problem of Sentiment Analysis (SA) has been studied well on the English language but not Arabic one. Two main approaches have been devised: corpus-based and lexicon-based.

110. Multiple Features: This dataset consists of features of handwritten numerals (`0'--`9') extracted from a collection of Dutch utility maps

111. Dorothea: DOROTHEA is a drug discovery dataset. Chemical compounds represented by structural molecular features must be classified as active (binding to thrombin) or inactive. This is one of 5 datasets of the NIPS 2003 feature selection challenge.

112. Steel Plates Faults: A dataset of steel plates’ faults, classified into 7 different types. The goal was to train machine learning for automatic pattern recognition.

113. Car Evaluation: Derived from simple hierarchical decision model, this database may be useful for testing constructive induction and structure discovery methods.

114. One-hundred plant species leaves data set: Sixteen samples of leaf each of one-hundred plant species. For each sample, a shape descriptor, fine scale margin and texture histogram are given.

115. Semeion Handwritten Digit: 1593 handwritten digits from around 80 persons were scanned, stretched in a rectangular box 16x16 in a gray scale of 256 values.

116. SECOM: Data from a semi-conductor manufacturing process

117. Amazon Commerce reviews set: The dataset is used for authorship identification in online Writeprint which is a new research field of pattern recognition.

118. Yeast: Predicting the Cellular Localization Sites of Proteins

119. Contraceptive Method Choice: Dataset is a subset of the 1987 National Indonesia Contraceptive Prevalence Survey.

120. REALDISP Activity Recognition Dataset: The REALDISP dataset is devised to evaluate techniques dealing with the effects of sensor displacement in wearable activity recognition as well as to benchmark general activity recognition algorithms

121. banknote authentication: Data were extracted from images that were taken for the evaluation of an authentication procedure for banknotes.

122. UJI Pen Characters: Data consists of written characters in a UNIPEN-like format

123. Diabetic Retinopathy Debrecen Data Set: This dataset contains features extracted from the Messidor image set to predict whether an image contains signs of diabetic retinopathy or not.

124. Mice Protein Expression: Expression levels of 77 proteins measured in the cerebral cortex of 8 classes of control and Down syndrome mice exposed to context fear conditioning, a task used to assess associative learning.

125. CNAE-9: This is a data set containing 1080 documents of free text business descriptions of Brazilian companies categorized into a subset of 9 categories

126. Geographical Original of Music: Instances in this dataset contain audio features extracted from 1059 wave files. The task associated with the data is to predict the geographical origin of music.

127. QSAR biodegradation: Data set containing values for 41 attributes (molecular descriptors) used to classify 1055 chemicals into 2 classes (ready and not ready biodegradable).

128. STUDENT ALCOHOL CONSUMPTION: The result also provides the correlation between alcohol usage and the social, gender and study time attributes for each student.

129. Parkinson Speech Dataset with Multiple Types of Sound Recordings: The training data belongs to 20 Parkinson's Disease (PD) patients and 20 healthy subjects. From all subjects, multiple types of sound recordings (26) are taken.

Supported By:

 In Collaboration With:

About  ||  Citation Policy  ||  Donation Policy  ||  Contact  ||  CML