LinkedIn 用機器學習提供雇主可能的職缺對象

先前看到「Learning Hiring Preferences: The AI Behind LinkedIn Jobs」這篇,LinkedIn 用機器學習提供雇主可能的對象。

依照官方的說法,這次提到的改進是透過雇主的行為調整推薦。當雇主對某個人有興趣的時候,LinkedIn 就會調整演算法去配合雇主有興趣的條件:

Based on how you interact with candidates, our algorithm learns your preferences and delivers increasingly relevant candidates across the Jobs product. If you’re consistently interested in candidates who are, say, accountants with leadership skills, or project managers who are adept at social media, we’ll send you more of those. And this all happens online in real time so that your feedback is taken instantly into account.

透過模擬 20% 的加成:

This new algorithm, which is used throughout the Jobs platform, performs nearly 20% better than the previous version in generating recommendations when we simulate our members' past hiring activity.

在 social network 這種演算法其實就是同溫層 (Echo chamber、Filter bubble),在 LinkedIn 這樣的行為不知道會不會牽扯到 Discrimination 的議題...

Percona XtraDB Cluster 5.6 的第一個 RC 版本...

Percona XtraDB Cluster 5.6 的第一個 RC 版本公告出來了:「Percona XtraDB Cluster 5.6.15-25.2 first Release Candidate is now available」。


Percona XtraDB Cluster now supports stream compression/decompression with new xtrabackup-sst compressor/decompressor options.

在「option compressor/decompressor」這邊可以看到這個設定,功能是在傳輸 SST 的過程壓縮。

看了範例設定,似乎也可以使用 gzip 以外的方式?bzip2 或是 xz 應該是可行的方案?

對跨機房之間的 full resync 應該是有幫助的... 像是 IPsec 的處理能力沒那麼高的時候 :p