or
or
By clicking below, you agree to our terms of service.
New to HackMD? Sign up
Syntax | Example | Reference | |
---|---|---|---|
# Header | Header | 基本排版 | |
- Unordered List |
|
||
1. Ordered List |
|
||
- [ ] Todo List |
|
||
> Blockquote | Blockquote |
||
**Bold font** | Bold font | ||
*Italics font* | Italics font | ||
~~Strikethrough~~ | |||
19^th^ | 19th | ||
H~2~O | H2O | ||
++Inserted text++ | Inserted text | ||
==Marked text== | Marked text | ||
[link text](https:// "title") | Link | ||
 | Image | ||
`Code` | Code |
在筆記中貼入程式碼 | |
```javascript var i = 0; ``` |
|
||
:smile: | ![]() |
Emoji list | |
{%youtube youtube_id %} | Externals | ||
$L^aT_eX$ | LaTeX | ||
:::info This is a alert area. ::: |
This is a alert area. |
On a scale of 0-10, how likely is it that you would recommend HackMD to your friends, family or business associates?
Please give us some advice and help us improve HackMD.
Do you want to remove this version name and description?
Syncing
xxxxxxxxxx
基於社群網路分析的交友軟體推薦系統
tags:
MLG
簡報網址:hackmd.io/@kuouu/MLG_final
提案簡報:hackmd.io/@kuouu/MLG_final_proposal
Introduction
AInimal 人工智慧社群養成
- The image file may be corrupted
- The server hosting the image is unavailable
- The image path is incorrect
- The image format is not supported
Learn More →網頁版 👉 ainimal.io
目前配對機制
- The image file may be corrupted
- The server hosting the image is unavailable
- The image path is incorrect
- The image format is not supported
Learn More →改良–基於個人特質的聊天活絡程度預測
Google Colab
- The image file may be corrupted
- The server hosting the image is unavailable
- The image path is incorrect
- The image format is not supported
Learn More →正確率
可能原因:
人格特質或許跟交友有關,但沒有辦法確定是“很像的人”還是“不像的人”容易成為朋友,甚至是情侶(大多數人用交友軟體的目的)
比起個性或興趣,我認為成為朋友的契機更注重於一些當下的因素,心情、場域、機緣等,因此決定使用社群網路來做推薦系統,並進行分析
配對成功?
會回訊息
聊得很熱絡
真命天子/女、一生的好朋友
軟體宗旨是找到真命天子/女,但得知後續發展可能要做調查,在訓練上會難以做標記,所以退一步到找到很會聊的人
Relative Work
Dating 2016 / 24 cites
Algorithm
程式碼 👉 github.com/erickuo5124/MLG_final
rule base
rule base
依據當下的平均聊天量由高至低做推薦
\[ 平均聊天量 = \frac{\sum_{聊天室}聊天字數}{聊天室數量} \]
期望結果
解決的問題
pseudo code
GNN
GNN
用 GCN 得到 node embedding ,並使用 similarity 做 user user 的 Collaborative Filtering
Experiment
Evaluation: recall
\[ Recall(pred, label) = \frac{pred 出現在 label\_list 出現次數}{label\_list 的長度} \]
Data set
分別為 2020, 2019 年 11/01~11/15 的資料
其實很多點“沒有往外的邊”,也就是很多使用者沒有在使用,可能只是下載來看一看就再也沒有登入了,因此在加入 user feature 之後,我將不活躍的使用者移除,並按照性別改變顏色,圖形變成如下
問題點
Conclusion
future work
END