Selama ini sebagian besar analisis dalam bidang pertanian, khususnya agribisnis dan sosial ekonomi pertanian menggunakan pendekatan ekonometrika dengan mendasarkan pada asumsi "linieritas". Pendekatan ini memberikan keunggulan dalam analisis ekonomi, seperti elastistas, return to scales, ataupun analisis fungsi permintaan dan penawaran dapat diketahui dengan mudah. Meskipun demikian, saat ini disadari bahwa permasalahan dalam bidang pertanian tidak sesederhana yang diasumsikan. Fungsi yang dianalisis mungkin merupakan fungsi polynomial. Beberapa variabel mungkin tidak dapat didefinisikan dengan jelas, khususnya variabel sosial. Oleh karena itu, penggunaan soft computing seperti model saraf buatan (artificial neural network), genetic algorithm, fuzzy logic mulai banyak dipergunakan untuk memecahkan berbagai persoalan yang bersifat non-linier, hazy dan subyektif. Berkaitan dengan hal tersebut, tulisan ini memberikan ilustrasi penggunaan model saraf buatan dalam bidang pertanian dengan mengambil kasus pada tanaman tebu. Perhitungan dilakukan dengan menggunakan multilayers networks sebanyak empat lapis dan proses belajar menggunakan algoritma back propagation. Proses pembelajaran dilakukan sampai terjadi overtraining untuk memetakan pola hubungan faktor penentu penerimaan dan pendapatan petani tebu. Hasil yang diperoleh menunjukkan bahwa biaya-biaya seperti sewa lahan, bibit, pestisida dan tenaga kerja memiliki kontribusi yang besar dalam mempengaruhi penerimaan dan pendapatan petani. Biaya sewa lahan yang tinggi dengan penggunaan biaya kerja yang rendah akan menekan pendapatan petani sehingga usaha tani tebu harus diusahakan secara intensif. Selain itu, biaya modal sendiri cenderung lebih menekan pendapatan dibandingkan dengan modal pinjaman.

The advancement of big data analytics is paving the way for knowledge creation based on very huge and unstructured data. Currently, information is scattered and growth tremendously, containing many information but difficult to be interpreted.Consequently, traditional approaches are no longer suitable for unstructured data but very rich in information. This situation is different from the role of previous information technology in which information is based on structured data, stored in the local storage, and in more advanced form, information can be retrieved through internet. Meanwhile, in Indonesia data are collected by many institutions with different measurement standard. The nature of the data collection is top-down, carried out by survey which is expensive yet unreliable and stored exclusively by respective institution. SIDeKa (Sistem Informasi Desa dan Kawasan/Village and Regional Information System), which are connected nationally, is proposed as a system of data collection in the village level and prepared by local people. Using SIDeKa, data reliability and readiness can be improved at the local level. The goals of the SIDeKa is not only local people have information in their hand such as poverty level, production, commodity price, the area of cultivated land, and the outbreak of diseases in their village, but also they have information from the neighboring villages or event at the national level. For government, data reliability will improve the policy effectiveness. This paper discusses the implementation and role of SIDeKa for knowledge creation in the village level, especially for the agricultural activities which has been initiated in 2015.

Due to the campaign supported by manufacturers of soy-oil in the United States which stated that consuming coconut oil might cause heart disease, coconuts’ popularity started to decline. As a result, the coconut was neglected by farmers and productivity was declining. Yet, recent study showed the opposite results. Coconut was good for health. As a result, the demand for coconut products increased. Indonesia, as the world’s largest coconut producers cannot maximize these opportunities due to aging coconut which have been neglected for many years. Coconut products may improve the income of farmers and encourage sustainable agriculture as well as diversify farmers’ income. Some simple methods, such as Klassen Typology or Location Quotient can be used to classify the potential area to be developed or rejuvenated. Nevertheless, these approaches are not able to generalize and not suitable to be implemented for cases with large data. This research tried to use cluster analysis such as dendrogram, Principle Component Analysis and Artificial Neural Network for classification. The results show that the dendrogram provided good results, whilst the Principle Component Analysis and Artificial Neural Networks required more data for better results.

No abstract

No abstract