About the data

Source

TOKEN2049 Week official side events sheet (CSV + PDF links). 351 events across 7 days (Mon, 5 Oct – Sun, 11 Oct), all in Singapore time.

Rows read

364

Valid events

351

Duplicates merged

5

Rows rejected

1

Field coverage

Share of events where the source provides each field.

Event name

100%

Date

100%

Start time

100%

End time

100%

Format

100%

Price

100%

Registration link

97%

Description

83%

Venue

73%

Host (from title)

87%

Named people

13%

Registration links come from the hyperlinks in the official sheet's PDF export, matched to each event row and checked against its date and times. Events the sheet marks “Invite Only” have no public link. Venues, descriptions and speaker lists aren't in the sheet; the data layer is ready to add them from event pages later.

Cleaning

• Duplicates. 5 listings that appear twice on the same day under the same name were merged: CoinFerenceX The Best Event Singapore - Day 1; CoinFerenceX The Best Event Singapore - Day 2; Sui Basecamp 2026 - Day 1; Arbitrum Founders Eve, Singapore 🇸🇬; Sui Basecamp 2026 - Day 2.
• Row 172 skipped. Looks like a test entry.
• Times. Events ending after midnight are moved to the next day. 5 events have end times that look unusual and are flagged on their pages.
• Titles. 11 titles are cut off at 75 characters in the source and shown with “…”.

How matching works

Topics are detected with transparent keyword rules on each listing — every tag keeps the words that triggered it. Your match score combines:

• Interest match 0–30Your interests vs. the event's topics
• Goal match 0–25Your goals vs. topics and format
• Role match 0–15What your role typically looks for
• Speaker relevance 0–10Named people or a talk format
• Networking 0–10How networking-oriented the format is
• Schedule fit 0–10On your days, no clash with your plan

Your “optimize for” choices re-weight these parts. The score says how well an event fits what you told us — not how good the event is.