Translation Induction on Indian Language Corpora Using Translingual Themes from Other Languages

Identifying translations from comparable corpora is a well-known problem with several applications, e.g. dictionary creation in resource-scarce languages. Scarcity of high quality corpora, especially in Indian languages, makes this problem hard, e.g. state-of-the-art techniques achieve a mean reciprocal rank (MRR) of 0.66 for English-Italian, and a mere 0.187 for Telugu-Kannada. There exist comparable corpora in many Indian languages with other ``auxiliary'' languages. We observe that translations have many topically related words in common in the auxiliary language. To model this, we define the notion of a translingual theme, a set of topically related words from auxiliary language corpora, and present a probabilistic framework for translation induction. Extensive experiments on 35 comparable corpora using English and French as auxiliary languages show that this approach can yield dramatic improvements in performance (e.g. MRR improves by 124% to 0.419 for Telugu-Kannada). A user study on WikiTSu, a system for cross-lingual Wikipedia title suggestion that uses our approach, shows a 20% improvement in the quality of titles suggested.

Microhyla laterite sp nov., A New Species of Microhyla Tschudi, 1838 (Amphibia: Anura: Microhylidae) from a Laterite Rock Formation in South West India

In recent times, several new species of amphibians have been described from India. Many of these discoveries are from biodiversity hotspots or from within protected areas. We undertook amphibian surveys in human dominated landscapes outside of protected areas in south western region of India between years 2013-2015. We encountered a new species of Microhyla which is described here as Microhyla laterite sp. nov. It was delimited using molecular, morphometric and bioacoustics comparisons. Microhyla laterite sp. nov. appears to be restricted to areas of the West coast of India dominated by laterite rock formations. The laterite rock formations date as far back as the Cretaceous-Tertiary boundary and are considered to be wastelands in-spite of their intriguing geological history. We identify knowledge gaps in our understanding of the genus Microhyla from the Indian subcontinent and suggest ways to bridge them.

El Nino and the Indian rainfall in June

We have addressed the question of whether the massive deficit of 42% in rainfall over the Indian region in June 2014 can be attributed primarily to the El Nino. We have shown that the variation of convection over the Northern part of the Tropical West Pacific (NWTP: 120-150E, 20-30N) plays a major role in determining the all-India rainfall in June with deficit (excess) in rainfall associated with enhancement (suppression) of convection over NWTP. In June 2014, the outgoing long wave radiation (OLR) anomaly over this region was unfavourable, whereas in June 2015, the OLR anomaly over NWTP was favourable and the all-India rainfall was 16% higher than the long-term average. We find that during El Nino, when the convection over the equatorial central Pacific intensifies, there is a high propensity for intensification of convection over NWTP. Thus, El Nino appears to have an impact on the rainfall over the Indian region via its impact on the convection over the West Pacific, particularly over NWTP. This occurred in June 2014, which suggests that the large deficit in June 2014, could be primarily attributed to the El Nino acting via intensification of convection over NWTP.

The 2014 M-w 6.1 Bay of Bengal, Indian Ocean, Earthquake: A Possible Association with the 85 degrees E Ridge

Oceanic intraplate earthquakes are known to occur either on active ridge-transform structures or by reactivation of their inactive counterparts, generally referred to as fossil ridges or transforms. The Indian Ocean, one of the most active oceanic intraplate regions, has generated large earthquakes associated with both these types of structures. The moderate earthquake that occurred on 21 May 2014 (M-w 6.1) in the northern Bay of Bengal followed an alternate mechanism, as it showed no clear association either with active or extinct ridge-transform structures. Its focal depth of >50 km is uncommon but not improbable, given the similar to 90 Ma age of the ocean floor with 12-km-thick overlying sediments. No tectonic features have been mapped in the near vicinity of its epicenter, the closest being the 85 degrees E ridge, located similar to 100 km to its west, hitherto regarded as seismically inactive. The few earthquakes that have occurred here in the past are clustered around its southern or northern limits, and a few are located midway, at around 10 degrees N. The 2014 earthquake, sourced close to the northern cluster, seems to be associated with a northwest-southeast-oriented fracture, located on the eastern flanks of the 85 degrees E ridge. If this causal association is possible, we believe that reactivation of fossil hotspot trails could be considered as another mechanism for oceanic intraplate seismicity.