The first topic is “Multilingual SMS-based Author Profiling: Data and Methods” published in Journal
of Natural Language Engineering (NLE), Cambridge. The collected corpus consists of SMS-based
profiles that contain text in Roman Urdu and English. The corpus is utilized for i) code-switching at
word-level and ii) empirical evaluation by applying existing monolingual author profiling methods.

The second topic is “Cross-lingual Single-document Abstractive Text Summarization for Science
Journalism”. Abstractive summarization task refers to generate human-like summaries automatically
that can be further used in different applications. Most existing research work focuses on monolingual
(abstractive and extractive) summarization while cross-lingual summarization is not thoroughly
explored. There are not much developed resources and methods for cross-lingual summarization. This
presentation covers my ongoing work on cross-lingual abstractive summarization
covering the research idea and plan.