Differences

This shows you the differences between two versions of the page.

--- volunteer_data:top [2022/06/13 23:43] – tim
+++ volunteer_data:top [2024/03/17 22:56] (current) – yang
@@ Line 1: / Line 1: @@
 ====== Volunteer Data Analysis ======
+====Volunteer Matching====
+  * [[volunteer_data:volunteer_matching| Project Page]]
 ====Persistent community detection====
@@ Line 6: / Line 11: @@
 ====Recommendation and social network analysis====
+  * [[volunteer_data:meetings| Meeting Notes]]
-**Meeting: Jan 17, 2022**
-  * Presented results of experiments on Volunteer data and Movielens
-  * Discussed the metrics for measuring performance on the two datasets
-  * Showed graphs of two methods and their performance on the two datasets
-  * Key: the BOA method still outperforms LDA in both datasets.
-  * **Comments and Future work**
-  * To compute metrics as a whole and not aggregate them separately for each user.
-  * To apply BOA method on a recommendation paper and compare the results
-  * To compile a representative code and send to the professor
-  * To read literature on latest work in RS.
-**Meeting: Jan 24, 2022**
-  * Discussed the possible alternatives of embedding the data, social net embedding
-  * Refined the life document to include user history pertaining to time, location and participation history
-  * Discussed a few papers that have used GNN for prediction
-  * **Comments and Future work**
-  * To read the papers related to Social Recommendation eg, Song et al (2019)
-  * Understand the nature of embeddings they used for the input data
-  * Summarise findings of papers (using GNN) whose data is similar to ours
-**Meeting: Feb 18, 2022**
-  * Presented a Survey on GNN Social recommendation
-  * Presented Item-to-Item KNN method used by Amazon
-  * Discussed two papers on Social Recommendation and presented their findings
-  * Paper 1: GNN for Social Recommendation - Fan, Ma,  Li, He,Zhao,Tang, and Yin. 2019. (ACM, WWW)
-  * Paper 2: Session-based Social Recommendation via Dynamic Graph Attention Networks, Song, Wang and Xiao (2019) – (ACM, WSDM)
-  * **Comments and Future work**
-  * To begin working on experiments related to dynamic session-based recommendation
-  * To make a list of the features and embeddings that can be used from our data,
-  * Determine session and neighbours based on our data, eg location based or participation based.
-**Meeting: March 1, 2022**
-  * **Attendees:**
-  * Professor Yang Li, Shutong and Tim (19:00-20:00)
-  * **Discussion**
-  * Tim presented a proposal on GNN
-  * **Comments and Future work**
-  * To clarify the research problem by formulating the underlying mathematical nitty gritties
-  * To consider geographic location and tasks types as input features and determine how to incorporate them
-**Meeting: March 23, 2022**
-  * **Attendees:**
-  * Professor Yang Li, Shutong and Tim (11:00-12:00)
-  * **Discussion**
-  * Shutong presented and discussed the plan on how the experiments will be conducted.
-  * Tim presented the findings on the first experiment for Session based GNN
-  * **Comments and Future work**
-  * Tim to write a brief methodology discussing his implementation of the algorithm
-  * Need for providing details for experimental plan and hyper-parameter tuning
-**Meeting: March 31, 2022**
-  * **Attendees:**
-  * Professor Yang Li, Shutong and Tim (11:00-12:00)
-  * **Discussion**
-  * Tim presented a write up of methodology section that also included experiment plan
-  * **Comments and Future work**
-  * Tim to give more clarity on the network architecture diagram
-  * To consider node embeddings as one hot key vectors
-  * To also include task type information on the volunteer preference
-  * Tim to remind Shutong and the professor every week before each meeting.
-  * Shutong suggested Tim to use PPT to draw flowchart and diagrams
-  * Shutong suggested to provide reference code for matching task types
-**Meeting: April 7, 2022**
-  * **Attendees:**
-  * Professor Yang Li, Shutong and Tim (11:00-12:00)
-  * **Discussion**
-  * Shutong presented her preliminary experimental results
-  * Tim discussed the DGRec framework.
-  * **Comments and Future work**
-  *Shutong to continue with her experimental plan
-  *Shutong to consider narrowing the research towards a more specific problem
-  * Tim to conduct experiment and reproduce the results in the paper
-  * **Attachments:**
-  * {{:volunteer_data:shutong_experiments.pptx}}
-  * {{:volunteer_data:graph_attention_network.docx}}
-**Meeting: April 14, 2022**
-  * **Attendees:**
-  * Professor Yang Li, Shutong and Tim (11:00-12:00)
-  * **Discussion**
-  * Shutong presented statistical data for Pioneers to be used in an article
-  * Tim presented the reproduced results from the paper
-  * **Comments and Future work**
-  *Shutong to write a summary of her latest findings
-  * Tim to consider a naive baseline approach that will be compared against the current method
-  * **Attachments:**
-  * {{ :volunteer_data:tim_progress_4.14.pptx |}}
-**Meeting: April 21, 2022**
-  * **Attendees:**
-  * Professor Yang Li, Shutong and Tim (11:00-12:00)
-  * **Discussion**
-  * Shutong presented a summary of her proposal and latest findings
-  * Tim compared the performance of voting process/UserKNN against the model of interest
-  * **Comments and Future work**
-  * Tim to explore at least one or two graph models for further comparison
-  * Shutong to make recommended adjustments to the presentation
-  * **Attachments:**
-  * {{ :volunteer_data:tim_progress_4.21pptx.pptx |}}
-**Meeting: May 16, 2022**
-  * **Attendees:**
-  * Professor Yang Li, Shutong and Tim (14:00-15:00)
-  * **Discussion**
-  * Shutong presented an updated version of her research findings, challenges and proposal. Greedy algorithm solves some of the common challenges in community detection.
-  * Shutong discussed latest findings on NCE and community detection for Futian and Nanshan districts
-  * Tim showed comparative results for DGRec and GR4URec. DGRec still outperforms baselines such as UserKNN
-  * **Comments and Future work**
-  * Shutong to determine the relationship between SHS and NCE results
-  * Shutong to summarize charts and send them to Pioneers in order to get feedback on changes that correspond to the detected time slots
-  * Shutong to make minor recommended adjustments to the presentation such as illustrating the important next steps as a pipeline instead of explaining them in text.
-  * Tim to analyze data to determine the optimal time window that gives dynamic sequences
-  * Tim to conduct experiment at district level, for instance using Futian and/or Nanshan data
-**Meeting: May 23, 2022**
-  * **Attendees:**
-  * Professor Yang Li, Shutong and Tim (14:00-15:00)
-  * **Discussion**
-  * Tim presented statistical information to determine appropriate time window
-  * Based on standard deviation and entropy, Futian and Nanshan had highest metrics at k=7 and k= 28, respectively.
-  * Shutong discussed the relationship between NCE, SHS and S-NCE
-  * Shutong reported that C-NCE has positive correlation with the number of tasks and volunteers, meaning that when the number of tasks/volunteers increase, the system becomes unstable. C-NCE also is positively correlated to the number of SHS [for more details, see attachment]
-  * **Comments and Future work**
-  * Shutong encountered an efficiency problem where its taking longer to compute results for Futian
-  * Shutong to show her computational method to pave way for alternative solutions
-  * Tim to consider double checking the data since the number of unique organisers keeps decreasing/constant as time window increases.
-  * Tim to conduct experiment at district level with induced structural breaks in the data
-  *  **Attachments**
-  * Shutong: {{ :volunteer_data:shutong_cnce.pptx |}}
-**Meeting: May 30, 2022**
-  * **Attendees:**
-  * Professor Yang Li, Shutong and Tim (14:00-15:00)
-  * **Discussion**
-  * Tim presented updated statistical information to determine appropriate time window. The average of Std and Entropy yielded a prediction window of 14 days. Structural breaks were used to determine breakpoints. DGRec with 14 day window was then trained on the breakpoints and results were presented for Futian district.
-  * Model performance during low volunteer turnout period is low, implying that when participation is low, volunteers behavior is not deterministic.
-  * Shutong discussed the meeting minutes from her previous presentation with another collaborating professor.
-  * **Comments and Future work**
-  * Tim to come up with hypothesis on factors that influence the participation of volunteers during low turnout period, ie, breakpoint 4.In the long run, to think about volunteer retention.
-  * Tim to extract weights from the model and analyze them
-  * Tim to keep doing experiments and compare the performance of DGRec with GR4URec
-  * Also report statistics of the total number of participants per each breakpoint.
-  * Instead of improving the model, think of how we can use it to analyze volunteer behaviour
-**Meeting: June 13, 2022**
-  * **Attendees:**
-  * Professor Yang Li, Shutong and Tim (14:00-15:00)
-  * **Discussion**
-  * Tim presented an analysis of hypotheses on volunteer participation. H1: The distribution of task types impacts daily volunteer participation, H2: Distribution of task types impacts model prediction.
-  * It was also noted that task types that appeared as 'noise' were not significant in OLS BP4 regression
-  * Shutong shared an outline of the experiments that she intends to do in the coming days/weeks.
-  * **Comments and Future work**
-  * Tim to construct hypotheses around improving model prediction accuracy and not sway far away from the main problem.
-  * Tim to consider directly testing if and how task types or volunteer participation impacts model prediction instead of applying indirect convolutions
-  * Tim to write a draft paper
-  * Shutong to follow up and execute the next tasks according to the experiment plan
-  * Shutong and Tim to work on PPTs for weekly group meeting scheduled for June 23rd