Humor is a unique and creative communicative behavior often displayed during social interactions. It is produced in a multimodal manner, through the usage of words (text), gestures (visual) and prosodic cues (acoustic). Understanding humor from these three modalities falls within boundaries of multimodal language; a recent research trend in natural language processing that models natural language as it happens in face-to-face communication. Although humor detection is an established research area in NLP, in a multimodal context it has been understudied. This paper presents a diverse multimodal dataset, called UR-FUNNY, to open the door to understanding multimodal language used in expressing humor. The dataset and accompanying studies, present a framework in multimodal humor detection for the natural language processing community. UR-FUNNY is publicly available for research.

A participatory stakeholder-based approach was employed to improve the quality of exportable fish and prawn landed in Kulierchar of Kishoregonj through reduction of the post harvest loss and its impacts were assessed. Primary data revealed that wet fish were minimally washed with clean water immediately after harvest. Ice boxes were not used in any stages from harvest to processing plants. Bamboo split baskets were mostly used to keep or carry fish. Landed fish were found to be dispersed on plastic mat or polythene paper kept on earthen floor of the wholesaler shed, called "Arat". Icing of individual catch soon after harvest was very insignificant, ranging from 2 -15 % only. Prawn were found to be iced during transport to landing center at little higher rate (32%), but most of the other fish were not significantly iced during transport to landing center. Primary awareness of the beneficiaries on quality deterioration of fish was assessed and the impacts of participatory awareness development tools (PRA and RRA, personal contacts, brain storming and awareness development training) on export, income, nutrition and improved handling of landed exportable prawn/fish were re-assessed. Efficacy of such measures to improve the situation were monitored for sixteen months. Awareness in terms of benefit of good practices was high in factory technicians. Other stakeholders had very shallow or poor knowledge on use of improved handling and processing of fish. Participatory tools employed in this study was found to be effective in developing awareness of stakeholders in different value chains of wet fish handling and distribution and reducing post-harvest loss.

No abstract

No abstract

The study was c o n d u c t e d to standardize t h e desire level of papaya (Carica papaya) latex on quality of cottage cheese prepared from buffalo milk. Cheese sample was prepared using whole milk collected from Lal Teer Livestock Development (BD) Ltd. Dairy farm. This experiment was conducted on five treatments named as sample BC2: buffalo control; A2: 5 drops (0.16g); B2: 6 drops (0.18g); C2: 7 drops (0.21g); and D2: 8 drop (0.24g). Overall score of cheese samples, B2 was highest (91.67±0.58) and A2 was lowest (77.67±9.29). Chemical analysis showed that, highest protein value 24.00±1.00 was found in B2 and lowest value 21.00±1.00 was in C2. On the other hand C2 content the maximum fat percent18.00±1.00 and A2 content the minimum fat percent 12.00±1.00. Results indicate that carbohydrate, ash and acidity have significant differences (p<0.01) but non-significant difference was found in total solids and moisture. Cheese from buffalo milk on coagulation time significantly different (p<0.01) but non-significant differences was found in yield.