Volume Time Series of Memetracker Phrases and Twitter Hashtags
Dataset information
Data contains the time series of the volume (the number of mention per hour) of 1,000 Memetracker phrases and 1,000 Twitter hashtags. Memetracker phrases are the 1,000 highest total volume phrases among 343 million phrases collected from Sep 2008 to Aug 2009. Twitter hashtags are the 1,000 highest total volume hashtags among 6 million hashtags from Jun to Dec 2009.
| Dataset statistics of each data set | |
|---|---|
| Number of time series | 1,000 |
| Length of time series | 128 |
| Time unit | 1 hour |
Source (citation)
- J. Yang, J. Leskovec. Temporal Variation in Online Media. ACM International Conference on Web Search and Data Mining (WSDM '11), 2011.
Files
| File | Description |
|---|---|
| MemePhr.txt | Memetracker phrases |
| TwtHtag.txt | Twitter hashtags |
Data format
Each item (a phrase or a hashtag) consists of two lines. The first line describes the meta-information about the item, and the second line shows the value of time series. Values are separated by tabs.
- 1:ID Content BeginTm NumQt PeakTm
- 2:Val(1) Val(2) ... Val(128)
Example:
5071230 public health... 2009-04-24 22:46:12 7 2009-04-26 18:16:12
20.980761 22.958107 24.416933 25.600956 ...
Description
- ID: ID of an item
- Content: The content of the phrase or the hashtag
- BeginTm: The time when the item is first mentioned
- NumQt: The number of slightly modified phrases that are derived from the phrase (only for Memetraker)
- PeakTm: The time when the item has its peak volume
- Val(i): The value of time series (Volume) at the i-th hour