Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Beyond words and actions: Exploring Multimodal Analytics and Collaboration in the Digital Age

Diego Miranda, Rene Noel, Jaime Godoy, Carlos Escobedo, Cristian Cechinel, Roberto Munoz

TL;DR

Results indicate that while planning poker doesn't significantly change total speaking or attention time, it leads to a more equitable speaking time distribution, highlighting its benefit in enhancing equitable team participation.

Abstract

This article explores Multimodal Analytics' use in assessing communication within agile software development, particularly through planning poker, to understand collaborative behavior. Multimodal Analytics examines verbal, paraverbal, and non-verbal communication, crucial for effective collaboration in software engineering, which demands efficient communication, cooperation, and coordination. The study focuses on how planning poker influences speaking time and attention among team members by utilizing advanced audiovisual data analysis technologies. Results indicate that while planning poker doesn't significantly change total speaking or attention time, it leads to a more equitable speaking time distribution, highlighting its benefit in enhancing equitable team participation. These findings emphasize planning poker's role in improving software team collaboration and suggest multimodal analytics' potential to explore new aspects of team communication. This research contributes to better understanding coordination techniques' impact in software development and team education, proposing future investigations into optimizing team collaboration and performance through alternative coordination techniques and multimodal analysis across different collaborative settings.

Beyond words and actions: Exploring Multimodal Analytics and Collaboration in the Digital Age

TL;DR

Results indicate that while planning poker doesn't significantly change total speaking or attention time, it leads to a more equitable speaking time distribution, highlighting its benefit in enhancing equitable team participation.

Abstract

This article explores Multimodal Analytics' use in assessing communication within agile software development, particularly through planning poker, to understand collaborative behavior. Multimodal Analytics examines verbal, paraverbal, and non-verbal communication, crucial for effective collaboration in software engineering, which demands efficient communication, cooperation, and coordination. The study focuses on how planning poker influences speaking time and attention among team members by utilizing advanced audiovisual data analysis technologies. Results indicate that while planning poker doesn't significantly change total speaking or attention time, it leads to a more equitable speaking time distribution, highlighting its benefit in enhancing equitable team participation. These findings emphasize planning poker's role in improving software team collaboration and suggest multimodal analytics' potential to explore new aspects of team communication. This research contributes to better understanding coordination techniques' impact in software development and team education, proposing future investigations into optimizing team collaboration and performance through alternative coordination techniques and multimodal analysis across different collaborative settings.
Paper Structure (4 sections, 8 equations, 5 figures, 3 tables)

This paper contains 4 sections, 8 equations, 5 figures, 3 tables.

Table of Contents

  1. Participant Detection and Tracking:
  2. Facial Landmark Detection:
  3. Accurate Face Direction Estimation:
  4. Detailed Analysis of Attention:

Figures (5)

  • Figure 1: A) Example of the facial points used to estimate its direction. B) Example of 2D points relative to the 3D points of the face. C) Diagram of the rotation axes $pitch$, $yaw$ y $roll$.
  • Figure 2: A) Example of a group activity viewed from the outside. B) Example of a group activity viewed with the 360-degree camera. The bounding box of the participant is indicated in green. The bounding box for the participant's face is indicated in red.
  • Figure 3: Heatmap of total speaking time in minutes per group and activity.
  • Figure 4: Boxplot for the Speaking Time Standard Deviation (STSD) Metric
  • Figure 5: Attention time chart for Group 11, Activity A and Activity B. The length of the arc segment represents the speaking time of each participant, the width of the arrows represents the attention time during speaking, and the direction of the arrow indicates the user's attention to others (arrows from the participant) and the time the participant receives attention from others (arrows towards the participant).