Sci & Tech

Persian Web Content Questioned

The spread of unoriginal Persian content has become a challenge to developing local search engines
The latest updates in the local search engines have focused on linking words with images.
The latest updates in the local search engines have focused on linking words with images.
People from different backgrounds need to be encouraged to employ their practical skills and scientific knowledge to create credible digital content.

Only 25% of 4 billion Persian web pages have original value and  the remaining 75% is either copied or spam content, says the executive director of Iranian-made search engine Parsijoo.

According to Amir Ali Kheyrandish, the new version of the search engine now covers 4 billion web pages in Persian. A year-on-year comparison shows that the number has almost doubled, ISNA reported.     

The issue that a great deal of unoriginal Persian content is produced, he said, poses a great challenge to the developers of Parsijoo.

"Sifting through the mass of irrelevant Persian web content is time-consuming and puts added pressure on the servers," he said.  

Furthermore, to maintain quality, Parsijoo's artificial intelligence, web-crawling, indexing and web ranking features must constantly be updated due to these problems, he said.

Search engine indexing collects, analyzes and stores data to facilitate fast and accurate information retrieval.

Web crawling is a program or automated script which browses the web in a methodical, automated manner. Search engines, use web crawling as a means of providing up-to-date data.     

Referring to the high level of irrelevant Persian web content, he said "this shows that content creation in the Persian language is not up to the mark."

Following global standards in the area of content creation "people from different backgrounds must be encouraged to employ their practical and scientific knowledge and create credible content with the vision to spread specialized information. This will also help those individuals to promote their own businesses."

"Parsijoo" (meaning Persian Search) is the second Iranian search engine; the first official Iranian search engine is 'Yooz'—Cheetah in Farsi.       

Several incentives are on offer to help promote the use of local search engines, state organizations have been told to integrate the use of local search engines in their official websites.

Despite this, locally developed search engines still remain largely unknown to Internet users in Iran.

In August 2016 an average of 500,000 users reportedly referred to the search engines daily, the head of Iran’s Communications and Information Technology Research Center, Alireza Yari said at the time.

"The quality of local search engines is poor and they cannot compete with global giants such as Google,” he concurred.

According to Parsijoo's ranking in Iran is currently 384 and Yooz comes way below at 1,604.

Detailing recent updates to the local search engines Yari recently said  that two new networks have been launched: a concept network known as Farsi Net and an image network Tasvir Net.

Farsi Net has been launched so that the search engines can better comprehend Farsi, this should minimize interpretation errors, he said.

Tasvir Net, on the other hand, aims to optimize image search.

"Together the two networks can forge links between words and images," he said.

In the first phase of the project which is underway at Iran Telecommunication Research Center six million photos have been linked to Persian words.

The number is set to increase to 14 million images in the second phase of the project while millions of Iranian images will also be added to show up in search results. 

Add new comment

Read our comment policy before posting your viewpoints