{"id":189,"date":"2024-03-26T15:11:58","date_gmt":"2024-03-26T15:11:58","guid":{"rendered":"https:\/\/education.ufl.edu\/sharkai\/?page_id=189"},"modified":"2024-04-23T17:40:13","modified_gmt":"2024-04-23T17:40:13","slug":"module-4","status":"publish","type":"page","link":"https:\/\/education.ufl.edu\/sharkai\/curriculum\/module-4\/","title":{"rendered":"Module 4"},"content":{"rendered":"

[et_pb_section fb_built=”1″ admin_label=”Header” _builder_version=”4.24.0″ _module_preset=”default” background_color=”#1C2F3D” use_background_color_gradient=”on” background_color_gradient_stops=”#041a3d 0%|rgba(25,69,136,0.24) 39%” background_image=”https:\/\/education.ufl.edu\/sharkai\/files\/2024\/03\/ai-20.png” background_position=”top_left” locked=”off” collapsed=”on” global_colors_info=”{}”][et_pb_row _builder_version=”4.19.2″ _module_preset=”default” max_width=”800px” global_colors_info=”{}”][et_pb_column type=”4_4″ _builder_version=”4.19.2″ _module_preset=”default” global_colors_info=”{}”][et_pb_text _builder_version=”4.24.0″ _module_preset=”c417ef0d-e927-4f7a-92d4-2634b80dd8a6″ link_font=”|800||on|||||” link_text_color=”#6FCAFC” link_font_size=”13px” link_letter_spacing=”1px” header_text_color=”#194588″ custom_margin=”||0px|||” locked=”off” global_colors_info=”{}”]<\/p>\n

Home<\/a> Curriculm<\/a>Module 4<\/a><\/p>\n

[\/et_pb_text][et_pb_text _builder_version=”4.24.0″ _module_preset=”_initial” header_font=”||||||||” header_text_color=”#FFFFFF” header_font_size=”50px” header_line_height=”1.2em” text_orientation=”center” custom_margin=”10px||10px||true|false” header_font_size_tablet=”40px” header_font_size_phone=”24px” header_font_size_last_edited=”on|desktop” global_colors_info=”{}”]<\/p>\n

Identifying Biases and Limitations in ML Datasets<\/h1>\n

[\/et_pb_text][\/et_pb_column][\/et_pb_row][\/et_pb_section][et_pb_section fb_built=”1″ admin_label=”FAQ” _builder_version=”4.24.0″ _module_preset=”default” background_color=”gcid-1bcf785a-50e1-437b-b09f-65567babc1de” locked=”off” collapsed=”on” global_colors_info=”{%22gcid-1bcf785a-50e1-437b-b09f-65567babc1de%22:%91%22background_color%22%93}”][et_pb_row _builder_version=”4.19.5″ _module_preset=”default” locked=”off” global_colors_info=”{}”][et_pb_column type=”4_4″ _builder_version=”4.19.2″ _module_preset=”default” global_colors_info=”{}”][et_pb_button button_url=”https:\/\/education.ufl.edu\/sharkai\/files\/2024\/04\/Module4.pdf” url_new_window=”on” button_text=”Download Module 4 PDF” button_alignment=”center” _builder_version=”4.24.0″ _module_preset=”default” custom_button=”on” button_text_size=”18px” button_text_color=”#1c2f3d” button_bg_color=”#EDBA75″ button_border_width=”2px” button_border_color=”#EDBA75″ button_border_radius=”80px” button_font=”|600|||||||” button_icon=”||divi||400″ button_on_hover=”off” custom_margin=”||40px||false|false” custom_padding=”12px|48px|12px|28px|true|false” hover_enabled=”0″ locked=”off” global_colors_info=”{}” button_bg_color__hover_enabled=”on|desktop” button_bg_color__hover=”RGBA(255,255,255,0)” button_bg_enable_color__hover=”on” button_text_color__hover_enabled=”on|hover” button_text_color__hover=”#EDBA75″ sticky_enabled=”0″][\/et_pb_button][et_pb_text _builder_version=”4.24.0″ _module_preset=”_initial” header_2_font=”|700|||||||” header_2_text_color=”rgba(255,255,255,0.83)” header_2_font_size=”36px” header_2_line_height=”1.2em” custom_margin=”||4px|||” header_2_font_size_tablet=”32px” header_2_font_size_phone=”20px” header_2_font_size_last_edited=”on|desktop” global_colors_info=”{}”]<\/p>\n

Module Overview<\/h2>\n

[\/et_pb_text][et_pb_text _builder_version=”4.24.0″ _module_preset=”default” text_font=”|600|||||||” text_text_color=”#ffffff” text_font_size=”15px” text_line_height=”1.8em” global_colors_info=”{}”]In this module, we will identify strategies for improving our machine learning models. To do this, we will look closely at the data used to train our machine learning models to identify potential biases. Biased training data is often due to an unintended pattern in the data. We will also utilize some of the more advanced features in Google\u2019s Teachable Machine to help us evaluate our models. Finally, we will recognize limitations of our machine learning models by showing that they cannot interpolate data when presented with broken teeth. [\/et_pb_text][\/et_pb_column][\/et_pb_row][et_pb_row _builder_version=”4.24.0″ _module_preset=”default” use_background_color_gradient=”on” background_color_gradient_direction=”128deg” background_color_gradient_stops=”rgba(31,126,152,0.43) 0%|rgba(28,47,61,0) 100%” custom_padding=”52px|36px|52px|36px|true|true” border_radii=”on|12px|12px|12px|12px” global_colors_info=”{}”][et_pb_column type=”4_4″ _builder_version=”4.24.0″ _module_preset=”default” global_colors_info=”{}”][et_pb_text _builder_version=”4.24.0″ _module_preset=”default” text_text_color=”#f7f7f7″ header_2_text_color=”#28BECC” header_3_text_color=”rgba(255,255,255,0.83)” global_colors_info=”{}”]<\/p>\n

Driving Question<\/h3>\n

How can we improve our model\u2019s performance?[\/et_pb_text][\/et_pb_column][\/et_pb_row][et_pb_row column_structure=”1_2,1_2″ _builder_version=”4.24.0″ _module_preset=”default” custom_margin=”24px||24px||true|false” global_colors_info=”{}”][et_pb_column type=”1_2″ _builder_version=”4.24.0″ _module_preset=”default” global_colors_info=”{}”][et_pb_text _builder_version=”4.24.0″ _module_preset=”default” text_font=”|600|||||||” text_text_color=”#ffffff” header_3_text_color=”#EDBA75″ global_colors_info=”{}”]<\/p>\n

Primary Learning Objectives<\/h3>\n